DEPARTEMEN PERMUKIMAN DAN PRASARANA WILAYAH

DIREKTORAT JENDERAL PRASARANA WILAYAH

JAVA ARTERIAL ROAD NETWORK STUDY

( J A R N S )
IBRD LOAN Number 3913 - IND

TECHNICAL REPORT No 10.

TECHNICAL REPORT No 10.

NETWORK MODEL DEVELOPMENT AND APPLICATION

October 2001
Ke

SU

MA
TE

RA

Bojonegara
Merak

JAKARTA

Cilegon

SERANG

[Link].

TANGERANG
BEKASI
Serpong

PANDEGLANG

Pamanukan

INDRAMAYU

KARAWANG

RANGKASBITUNG

Labuhan

Jangga

Narogong
Cileungsi

Lohbener

Cikampek

Karangampel
Jatibarang

Sadang

Jatiluhur

Kanci

Cipeo

PURWAKARTA

BOGOR

JEPARA

SUBANG

Ciawi

Dawuan

CIANJUR

[Link]

PATI
KUDUS

Kadipaten

BANDUNG
Citatah

MAJALENGKA

BREBES

Losari

Tanjungsari

TEGAL

PEMALANG

Trengguli

PEKALONGAN

Pejagan

SUMEDANG

SUKABUMI

REMBANG

CIREBON

Palimanan

Cijelag

SEMARANG

KENDAL

BULU

JATI

DEMAK

TUBAN

Ketapang

Sotobar

BATANG

BLORA

Cileunyi
SLAWI
Nagreg

Kaliwungu

KUNINGAN

BANGKALAN
Babat

PRUPUK

LAMONGAN

PURWODADI

GRESIK

Widang

UNGARAN

Kamal
PAMEKASAN

BOJONEGORO

GARUT

Bawen

SAMPANG

SURABAYA

Baturaden

TASIKMALAYA

WONOSOBO
CIAMIS

[Link]

PURBALINGGA
PURWOKERTO

SUMENEP

P. MADURA

GODONG

Waleri

BANJARNEGARA

Mantingan

SALATIGA

TEMANGGUNG

Waru

NGAWI

MOJOKERTO

SRAGEN
MAGELANG

Mojosari

Kartosuro

JOMBANG
BOYOLALI

Gemekan

Caruban

KR ANYAR

SURAKARTA

MADIUN

NGANJUK

SIDOARJO

Gempol
Pandaan

PASURUAN

Kertosono

SUKOHARJO
KEBUMEN

CILACAP

PURWOREJO

PANARUKAN

Pilang

MAGETAN

SLEMAN

Tretes

KLATEN

PROBOLINGGO

Kejayan

WONOGIRI

YOGYAKARTA

PONOROGO

KEDIRI

BANTUL

BONDOWOSO

MALANG

WONOSARI

Gilimanuk

BLITAR

TRENGGALEK

LUMAJANG
TULUNGAGUNG

JEMBER

Ketapang

Cekik

BANYUWANGI
PACITAN
Pengambengan

Carl Bro International a|s

In association with

MVA Asia Ltd.

PT. Amythas
PT. Jasa Mitra Manunggal

ND. Lea Consultants Ltd.

PT. Dressa Cipta Rekayasa
PT. Multi Phi Beta

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

TECHNICAL REPORT NO. 10:

NETWORK MODEL DEVELOPMENT AND APPLICATION

TABLE OF CONTENTS
INTRODUCTION

1.

1.1

Objectives

1.2

Background

1.3

Scope of this Report

2.

JARNS NETWORK MODEL OVERVIEW

2.1

Introduction

2.2

Study Area and Zone System

2.3

Vehicle Types Modelled

2.4
The Study Area Networks
2.4.1
Development of Base Year Network
2.4.2
Development of Future Networks
2.4.3
Other Model Parameters
2.4.4
Network Model Outputs and Evaluation
2.4.5
JARNS Modelling Software

4
4
5
6
6
7

3.

TRAFFIC SURVEYS AND DATA COLLECTION

3.1

Introduction

3.2
Traffic Surveys
3.2.1
Other Traffic Data Sources
3.2.2
Other Surveys and Data

8
8
8

3.3
Scope of Traffic Surveys
3.3.1
Survey Site Selection
3.3.2
Survey Time Periods and Coverage

9
9
9

3.4
Traffic Count Data
3.4.1
JARNS Traffic Count Surveys
Source: JARNS Traffic Surveys

3.4.2

Traffic Count Data from Other Sources

3.5
Roadside Interview Surveys (RIS)
3.5.1
General Background
3.5.2
Car O-D Surveys

Carl Bro International a|s and Associates

9
9
11
11
11
11
12

Java Arterial Road Network Study (JARNS)

3.5.3
3.5.4
3.5.5
4.

Technical ReportNo.10:
Network Model Development and Application

Goods Vehicles O-D Surveys

Angkot O-D Surveys
Bus O-D surveys

DEVELOPMENT OF BASE YEAR VEHICLE TRIP MATRICES

12
13
13
15

4.1

Introduction

15

4.2

Development of Observed Trip Matrices

15

4.3
Synthesis of Base Year Trip Matrices
4.3.1
Build Prior Trip Matrices
4.3.2
Initial Matrix Estimation
4.3.3
Final Matrix Estimation

17
17
18
19

5.

21

5.1

JARNS NETWORK MODEL DEVELOPMENT AND VALIDATION

Introduction

21

5.2
Network Data Sources and Coding
5.2.1
Network Model Data Sources
5.2.2
Coding of Study Area Network General
5.2.3
Coding of Strategic Network Links
5.2.4
Coding of Urban Area Links
5.2.5
Coding of Urban Area Corridors
5.2.6
Toll Roads and Toll Road Access Links
5.2.7
Rural Area Roads
5.2.8
Summary

21
21
21
23
26
27
27
27
27

5.3

Link Capacity and Speed Flow Relationships

28

5.4
Base Year Assignment and Model Validation
5.4.1
Assignment Time Period and PCE factors
5.4.2
Assignment Results Validation

29
30
31

6.

33

FUTURE TRAVEL DEMAND

6.1

Introduction

33

6.2

Trip End Models

33

6.3

Future Travel Demand Matrices

35

6.4

Forecast Future Pre-Load Traffic Volumes

35

7.
7.1

NETWORK MODEL APPLICATION

Introduction

7.2
Development of Future Year Networks
7.2.1
Base 2005 Network
7.2.2
Future Year 2010 Networks
7.2.3
Future Year 2020 Networks

Carl Bro International a|s and Associates

36
36
36
36
37
37

ii

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

7.3

Future Year Network Model Parameters

38

7.4

Testing of Future Year Network Scenarios and Evaluation

38

7.5

Network Model Inputs to the Economic Evaluation Framework

38

APPENDICES:
APPENDIX A:
APPENDIX B:
APPENDIX C:
APPENDIX D:
APPENDIX E:
APPENDIX F:
APPENDIX G:
APPENDIX H:

ZONE SYSTEM DEFINITION

TRAFFIC SURVEY DATA
MATRIX ESTIMATION
NETWORK DATABASE AND CODING
ROAD CAPACITY AND SPEED FLOW
RELATIONSHIPS
ASSIGNMENT VALIDATION
TRIP END MODELS
TRAFFIC GROWTH FACTORS

Carl Bro International a|s and Associates

iii

Java Arterial Road Network Study (JARNS)

1.
1.1

Technical ReportNo.10:
Network Model Development and Application

INTRODUCTION
Objectives

A key component of the JARNS study was to develop a road network model capable of
providing road traffic estimates for the strategic road network of Java for the next twenty
years. The model needs to be responsive to development changes and robust enough to
provide the testing and evaluation of alternative network development strategies, involving
both toll and non-toll road capacity expansion programmes. This report provides a full
documentation on the JARNS network model development and its application. A summary
report on network model has also been included in the JARNS Final Report as Appendix C.

1.2

Background

The JARNS study included the development of a network model based on the use of existing
data where possible and by carrying out new traffic surveys where appropriate. As a first
step, a review of the existing data sources was undertaken and reported in the Inception
report.
There exists a considerable amount of data on the Java road network within the Indonesian
Road Management System (IRMS) and the Automated Road Monitoring Study (ARMS)
database systems. For data on traffic itself, the data within the IRMS database and from other
sources was limited, and lacking in detail and accuracy. These issues were noted earlier, and
were fully addressed during the JARNS surveys and data collection program.
A review was undertaken of road traffic demand forecast models developed for studies
carried out in Java. This revealed that most of these models are limited in scope, and were
largely developed for specific geographical areas within Java, or for checking the feasibility
of particular road / toll road schemes. The major highway projects that covered a wider area
of Java were the CAPEX and North Java Corridor study, but models and data from these are
now of limited relevance.
The most recent study that looked at the road traffic demand at a strategic level was the
Transport Sector Strategy Study (TSSS). The traffic model developed for the TSSS was also
reviewed. The study only modelled inter-kabupaten road traffic in Java, and at an aggregate
level for all vehicle types. This is considered too coarse for the level of detail required for the
JARNS scope of analysis. The reported results of growth in traffic demand were also of
limited use, as the TSSS forecast time frame was limited to the next ten years. However,
where appropriate information and data from the TSSS study was used in the development of
the JARNS network model.

1.3

Scope of this Report

This Section comprises of seven sections. In the next section an overview of the JARNS
network model is presented. Section 3 gives a brief description and summary of survey
results. Sections 4 and 5 present the development and the validation of the trip matrices and
networks. Section 6 illustrates the forecast methodology and a summary of results, followed
by descriptions of forecast year networks and model application in Section 7. A Full
discussion of the output model results for alternative network development strategies and
scenarios is given in Section 6 of the JARNS Final Report.
There are eight Appendices to this report, which provide further details on the model
development and application.

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

2.
2.1

Technical ReportNo.10:
Network Model Development and Application

JARNS NETWORK MODEL OVERVIEW

Introduction

The JARNS study developed a simple but robust model for forecasting of traffic on the
strategic road network of Java. The traffic model is fully integrated with the JARNS Socioeconomic model SPADEL (see Technical Appendix A, and Technical Report No. 9), from
which it takes the alternative planning scenario inputs. The network model outputs are then
used as inputs to the evaluation framework for testing / evaluation of alternative road network
development strategies. The overall structure of the network model is shown in Figure 2.1.
This Section describes the key elements of the model, which need to be specified at the
outset, and these elements form the basis of the whole model.

2.2

Study Area and Zone System

The study area comprises the whole of the islands of Java and Madura. The two neighbouring
islands of Bali and Sumatra are represented as external zones. For road traffic studies, the
study area is divided into smaller units called traffic zones. The traffic demand model then
represents travel (number of trips) in or between these areas. The size and boundaries of the
traffic zones are defined in such a way as to capture the level of detail required by the study
objectives. For example, for an intra-urban travel study the urban area will be divided into
smaller units to reflect travel within the urban area, and areas around will be coarsely
aggregated to only represent travel to/from or through the urban area. In the case of an interurban study such as JARNS, the urban areas are best represented at a coarse level.
As the study objective was to look at inter-urban travel only, the study area zone system was
designed to capture the maximum amount of inter-urban traffic and to exclude intra-urban
local traffic, where possible. The size of each zone was determined such that each contains
about the same level of population, with the exception of major metropolitan and urban areas.
These were represented at an aggregate level of Kotamadya. All zone boundaries conform to
the administrative boundary of Kabupaten/ Kotamadya, and where a Kabupaten is divided
into many zones, the zone boundaries conform to the smaller administrative unit of
Kecamatan.
The JARNS study area was divided into 264 such traffic zones, 262 for Java, and Bali and
Sumatra represented by 2 external zones. The zones were mapped using the combination of
Kecamatan boundary mapping and a full description is given in Appendix A.

2.3

Vehicle Types Modelled

In Indonesia, 11 types of motorised road vehicles are usually defined for surveys. These are:
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.

Motorcycles,
Private Car/Jeep/Station Wagons or other such passenger vehicles
Small vans used as public passenger vehicles (Angkot)
Light Goods Vehicle (Pick-up trucks / Utility vehicles)
Small Bus
Large Bus
Trucks with 2-axles and 4-wheels
Trucks with 2-axles and 6-wheels
Rigid 3-axle Trucks
Truck Trailer
Truck and semi-trailer

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Figure 2.1: Overview of JARNS Network Model

A R M S / IR M S
D ata base

O th er Stu dies
D ata bases

R oa d O /D
Su rvey s

Traffic C ou nt
D ata bases

D ev elop N etw ork M od el &

Spe ed Flo w R elation sh ips

Partially O bserv ed
Trip M atrces

JA R N S N etw o rk
M o del D ataba se

JA R N S M atrix
Estim ation by
V eh icle T ype

B ase Y ear M od el V alid ation

B ase Y ear
V eh icle T rip
M a trices

B ase Y ear
N etw ork

JA R N S SPA D E L
M o del Socio eco nom ic D a tabase
B as e Y ear D ata

D ev elop Tra vel D em an d

M o dels a nd C alibra te
Param eters

D ev elop B ase N etw ork

In clud ing O ngo ing /
C om m itte d R oa d
Sch em es

F u tu re Y ea rs D ata

Forecast F uture Y ears

V eh icle T rip M atric es

A ssig n Fu ture D em a nd to N etw ork Stra tegies

Ev aluation Fram ew ork

Ev aluate Test S trateg ies / Sc enarios
C he ck N e tw ork O p eratio nal E fficie ntly and
A ssess Policy / Stra tegy O ptio ns

Ev aluate N etw o rk
Stra tegies / Sche m es

No

A v ailable
N etw ork U pg rad e
R eso urces

Is S trateg y
A cc eptab le

Y es
Pre pare
Im plem entatio n Plans

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

The traffic survey analysis showed that during the surveys it was difficult to distinguish
between small and large buses, and in many instances there are buses of many different sizes.
Most small buses also serve passengers for inter-urban travel. They were also observed to
carry similar numbers of passengers to the large buses, and their trip-length frequency
distribution was found to be similar to the large buses. Hence, the data for the two types of
buses were combined to form a single vehicle category Bus for the public passenger
vehicles. Angkots operate a frequent service over short distances on fixed routes; therefore,
these were retained as a separate category.
The trip length analysis of vehicle types 7 and 8 (2-Axle Trucks) revealed that the two types
perform similar functions of distribution of goods over short to medium distances. The total
numbers of each vehicle type observed were small. Therefore, for modelling purposes these
two types of vehicles were combined to form a single category of Medium Goods Vehicles
(MGV). Vehicle type 4 (Utility vehicles) were retained as separate category as Light Goods
Vehicles (LGV). The numbers of multi-axle (3 or more) trucks (vehicle types 9, 10 & 11)
were observed to be small, and therefore did not warrant to be modelled as separate
categories. These three types of multi-axle trucks were modelled together, as a single
category called Heavy Goods Vehicle (HGV).
Travel Demand Matrices of zone-to-zone travel were developed for the following vehicles
categories:

Car (vehicle type 2)

Light Goods Vehicle LGV (vehicle type 4)
Medium Goods Vehicles MGV (vehicle types 7 & 8)
Heavy Goods Vehicles HGV (vehicle types 9, 10 & 11)
Buses (vehicle types 5 & 6)

Motorcycles and Angkots

No trip matrices were developed for travel by motorcycle or Angkots. As stated above, these
vehicles usually make short distance intra-zonal trips along a route. Their effect on the road
capacity was taken into account by including them as pre-loads on the network, in the
assignment model.
The JARNS traffic surveys recorded the numbers of non-motorised vehicles. The survey data
showed that numbers of such vehicles were very low on the inter-urban roads. Hence, these
vehicles were excluded from any further analysis in this study.

2.4

The Study Area Networks

2.4.1

Development of Base Year Network

The roads to be included in the network model, was dictated by the strategic nature of the
study. The base year strategic network has been described in detail in Section 2.5 of the main
report. The strategic network includes:

All inter-urban toll roads, and urban toll roads that act as major bypasses for the
main urban areas, such as those around Cirebon, Bandung and Semarang.
Urban toll roads of Jakarta and Surabaya were also included in the network to
act as the main arteries /corridors serving these metropolis, provide a direct link
to the centre, and these also serving as a through route for the non-local traffic.
In total, there were some 439 km of toll roads in the JARNS network model.

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Of the 2,760 km of arterial roads in the JARNS study area, 2,539 km were included in the
strategic road traffic model network. The remainder of the arterial roads were either part of
the urban areas, or were some isolated sections. These sections of the arterial roads were
considered inappropriate to be part of an inter-urban strategic network.
Similarly, some 1,337 km (96%) of primary collector (K1) roads were included in the
strategic network. In order to provide direct routes from most of the main towns and urban
areas of Java to the main section of the strategic network, 3,496 km of the remainder of the
collector roads were included in the strategic network to provide collector/distribution
functions.
In addition to the above strategic road network, the JARNS model network includes a number
of urban areas roads and notional links. The main function of these roads/links in the study
area network to be simulated, was to cater for the movement of strategic traffic in and out of
the urban areas, and to allow for the passage of through inter-urban traffic. Another important
function of these urban area roads/links was to adequately represent the effect of congestion
on the inter-urban travel times. While JARNS cannot attempt to model urban traffic, it was
nevertheless essential to adequately represent the urban area networks. The modelling of
these links/roads was therefore not to the same level of detail as the strategic network. A brief
description and their scope is outlined below, further details on their full role in the network
model are given in Section 5 of this Appendix.

2.4.2

Urban / Kota links: These links are part of the national roads that extend into
urban areas, and have been designated as urban links by the IRMS/ARMS
database systems. These links lie within urban areas along the strategic routes,
and serve both the local and through traffic. In the JARNS, network model there
are some 414 km of such roads.
Urban Area Corridor Links: These are notional links, which represent several
urban area streets and roads used by the local and through traffic. These links
were necessary for the network model to complete the network. In the JARNS,
network model there are 182 km of such links. The majority of these are in
Jakarta, (153 km) and remainder in major towns of West and Central Java
provinces.
Toll Road Access Links: These are sections of roads, which provide connections
to the toll roads from the adjacent network. These links form an essential part of
the network, and represent 67km of roads.
Other Rural Area Links: There is 602km of rural roads included in the JARNS
network model. These roads provide north-south connections to the strategic
network for the rural population of southern seaboard of west Java. These roads
carry little or nor strategic traffic, and therefore were not considered as part
JARNS strategic network development options. It was assumed that any upgrade
of these roads would be the responsibility of the local area authorities, like the
development and upgrade of other Kabupaten/Kecamatan roads.
Development of Future Networks

The future year networks developed for testing alternative network scenarios / strategies were
based on the 2005 network. This 2005 network was developed from the base year (2000)
network by including the following new road schemes and upgrades:

All road schemes (bypass or upgrades) currently under construction;

All road improvement schemes for which funds are committed, and to be built
within the next five years; and
All planned and committed road schemes, construction of which would be likely
to be completed by the year 2005.

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

This provided the base network for the development and testing of the forecast year (2010
and 2020) strategies and scenarios. The development of future network strategies / Scenarios
are discussed in Section 5 and 6 of the main report, and the testing process is described in
Section 7 of this Appendix.
2.4.3

Other Model Parameters

Time Periods Modelled

All travel demand matrices were developed for 24-hour AADT. For the network assignment
model, the 24-hour vehicle trip matrices were converted to a typical one-hour travel matrix,
including a proportion of trips from each of the five vehicle category matrices. Motorcycle
and Angkot pre-loads were also converted to one-hour volumes. The typical hour adopted for
the study represented traffic observed during an average hour of the day light hours ([Link]), the 12-hour survey duration. The analysis of the survey data showed that the
distribution of traffic during the day varies a little for most vehicle categories, but that there is
no pronounced peak. See Figure 5.7.
The typical average hourly factors adopted for each vehicle category were derived from the
survey data and these factors varied between 4.9% for the buses to 6.6% for the motorcycles.
It should be noted that morning or evening peak period modelling is usually carried out for
urban studies. For inter-urban travel, a distinct peak is not very relevant as most trips are over
long distances, and may stretch over many peak periods.
Passenger Car Equivalent Factors
All trip matrices were converted to the common unit of Passenger Car Units using the PCE
factors specified in the Indonesian Highway Capacity Manual (IHCM).
Speed Flow Relationships and Assignment Parameters
Speed flow relationships are crucial to any capacity restraint (congested network) assignment
modelling. In Indonesia, the IHCM provided a sound database for the development of speed
flow relationships. The main source of road network data for this exercise was obtained from
the IRMS/ARMS databases and the ARMS digital images.
The choice of path in the assignment process is based on the Generalised Cost (GC) of travel,
which includes:

Value of travel time;

Perceived cost of travel related to the distance travelled;
Perceived value of tolls paid as out-of-pocket costs.

These parameters differ for different types of vehicles. Average values were estimated
reflecting the vehicle mix in the average peak hour traffic stream. Details of these parameters
are given in Section 5.
2.4.4

Network Model Outputs and Evaluation

The JARNS network model provides a variety of results. The main model output is the
network assignment results, which estimate link traffic volumes for the typical hour
modelled, and in terms of AADT. The model assignment results used for the operational
evaluation of the network in terms of Volume Capacity Ratio (VCR). The VCR could be
viewed graphically or output as link based tables to be used for further analysis. Network

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

travel summaries in terms of vehicle-kilometres and vehicle-hours can be calculated. These

statistics were direct input to the network evaluation framework that computes the benefits
(as difference between a base case and scheme test) of a scenario / scheme accrued over the
whole network.
An accident cost estimation module was used, which requires network statistics of amount of
travel by road, by a separate classification of roads. Network travel statistics of vehicle-hours
travelled by motorcycles and Angkot were also estimated and input to the overall evaluation
framework.
The model was also used to produce network wide zone-to-zone travel time, distance and
Vehicle Operating Cost (VOC) skim matrices for the base case, and test scenario. These skim
matrices were based on paths built using the converged assignment network and generalised
cost of travel as the route choice parameter. These matrices were then directly used to
calculate the economic benefits or dis-benefits weighted by the travel demand (number of
trips between each zone pair). The sum total of all these trip-weighted time-savings, distance
related and VOC savings represented the total benefits for the scheme, which were then used
by the economic evaluation framework.
2.4.5

JARNS Modelling Software

The JARNS network model is simple in nature, but it is robust and detailed enough to fully
achieve the objectives of the study. The model is developed within the MVAs own transport
planning suite of programs, the TRIPS package.
For survey data analysis and other similar tasks Microsoft Access and worksheet package
EXCEL has been used. SPSS for Windows was used for statistical analysis, and trip-end
model development. The Map-Info package was used for plotting of networks and study area
maps and zonal information.

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

3.

Technical ReportNo.10:
Network Model Development and Application

TRAFFIC SURVEYS AND DATA COLLECTION

3.1

Introduction

The main data required for modelling was the necessary Origin-Destination (O-D) patterns of
traffic in Java. It is these O-D patterns that provide the information on the amount and
distribution of travel, and hence the cost of travel on the network. These Roadside Interview
Surveys (RIS) require considerable resources, and are complex to analyse. Therefore could
only be conducted at a limited number of sites. At any site, a sample of the total traffic
passing is interviewed to obtain the O-D and other travel related information. In order to
expand the sample data to represent total traffic, it is also necessary to count total traffic
passing that site. These total traffic count surveys, usually called Manual Classified Counts
(MCC) are conducted simultaneously at the RIS sites. These provide the data needed to
gross-up the sample O-D survey results. Readers interested in further details are referred to
the Technical Report 2, Traffic Surveys and Data Collection. However, some summary
results are given in Appendix B.

3.2

Traffic Surveys

The RIS survey provided some of the O-D travel patterns throughout Java giving partially
observed trip matrices. The modelling approach adopted is reliant on the method of Matrix
Estimation, which estimates the remainder of the unobserved O-D movements. This process
requires collection of MCC data at numerous other key locations across the network. The
MCC data was also used to update the previously-collected count data, available from
numerous other sources. The JARNS study carried out a comprehensive set of RIS and MCC
surveys, in total at one hundred sites in Java. The remainder of this section briefly describes
the conduct, analyses, and summary results obtained from these surveys.
3.2.1

Other Traffic Data Sources

Traffic data was available from a number of other sources. It was checked for consistency,
and where appropriate was used in model development, validation and calibration. The major
sources of traffic data were:

3.2.2

IRMS database
ARMS survey data including moving observer count data
Toll roads and other parallel roads traffic count data from Jasa Marga
Automatic Traffic Count (ATC) data from West, Central and East Java sources
TSSS study total vehicle trip matrices and summary of count data
Traffic Count data from the on-going SURIP studies
Road Capacity Expansion Project Phase II (CAPEX-2) data
Madura-Surabaya, Java-Sumatra, and Java-Bali vehicular ferry traffic count data
were obtained from the appropriate authorities.
Other Surveys and Data

In addition to the traffic data collection exercise, Value of Time (VOT) and a limited number
of informal journey time surveys were also conducted. The work on value of time surveys
has been reported in Technical Report 7.

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

3.3

Technical ReportNo.10:
Network Model Development and Application

Scope of Traffic Surveys

3.3.1

Survey Site Selection

The Traffic survey programme had two main components: RIS and MCC data collection. The
selection of RIS (O-D) sites was based on the review of major road transport corridors in
Java. This ensured that all long distance trips are covered, and the short distance local trips
are not over represented. MCC sites were selected in such a way that all major east-west and
north south routes, and links into and out of major cites were fully covered. All survey sites
were located away from the main towns, to capture the maximum amount of inter-urban
traffic, and to minimise the inclusion of local intra-urban traffic.
Neither RIS nor MCC surveys were conducted on toll roads, because this was operationally
difficult, and toll road traffic count data was available for most mid-link and on/off ramps.
Traffic surveys were conducted at 100 stations across Java. The exact RIS and MCC survey
locations and the full survey programme are detailed in Technical Report 2, Traffic Surveys
and Data Collection. However, some summary results are presented in Appendix B.
3.3.2

Survey Time Periods and Coverage

Traffic surveys (RIS/MCC) were conducted either for 12-hour or 24-hour periods. The 12hour period stretched over the day light hours between 06:00 to 18:00. The 24-hour period
stretched from 06:00 hour to 06:00 the following day. Both 12-hour and 24-hour surveys
were conducted from Monday to Fridays. No surveys were carried out during Saturdays and
Sundays. Public Holidays were also avoided. The Table 3.1 below gives the extent of the
traffic survey coverage over the study area.
Table 3.1: Number of Traffic Sites in Java by Type and Duration of Survey Period
Survey Type
RIS Surveys
Sub-total RIS Sites
MCC Surveys

Survey Period

West Java

12-Hours

Central
Java
12

24-Hours

12/24 Hours

East Java

Total

30

11

14

12

37

12-Hours

20

25

21

66

24-Hours

12

10

12

34

32

35

33

100

Total Sites
12/24 Hours
Note: MCC Sites Include the RIS Survey Sites

3.4
3.4.1

Traffic Count Data

JARNS Traffic Count Surveys

Traffic count surveys involved continuously counting all traffic by predefined types of
vehicles passing the survey locations. The vehicle classification was based on the IRMS
classification of vehicles, to ensure compatibility with the other count data sources.
The number of vehicles observed (2-way) at each site varied from less than 5,000 vehicles
per day to just over 100,000 vehicles at a site (Waru - Sidoarjo) just outside Surabaya.
However, the survey sites represented a broad range of traffic volumes observed at locations
in all three provinces. The Table 3.2 below shows the range of 2-way traffic volumes
observed in each province.

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Table 3.2: Number of Sites by Range of Traffic Volumes Observed, by Province

Range of 2-Way Daily
Traffic (AADT)
Less than 10,000

West Java

Central Java

East Java

10

Total
All Sites
19

10,000 15,000

21

15,000 20,000

18

20,000 25,000

16

25,000 30,000

13

30,000 40,000

Over 40,000

35

33

100

Total
32
Source: JARNS Surveys (AADT Includes Motorcycles)

It can be seen that on inter-urban roads the traffic volumes are generally low, only at 13
locations the 2-way observed volumes were in excess of 30,000 vehicles per day. At nineteen
locations, the total daily traffic did not exceed 10,000 vehicles. Only at seven sites the traffic
volumes were over 40,000 vehicles per day, and at four of these locations the numbers of
vehicles were less than 50,000.
Analysis of traffic composition showed that by far the majority of vehicles are motorcycles,
accounting for over 35% of all vehicles observed. The proportion of motorcycles in East Java
was highest, close to 45% of all traffic. Distribution of cars was higher in West Java than
Central and East Java. Angkots accounted for over 21% all traffic in West Java compared to
just over 7% in Central and East Java provinces. This shows that the choice of mode is
somewhat similar in Central and East Java, i.e. majority of the short distance travel is by
motorcycles. The proportion of cars observed at sites in West Java was close to a quarter of
all traffic and slightly lower (21-22%) in Central and East Java. The proportion of buses and
goods vehicles was almost similar across three provinces. This variation in the distribution of
traffic across provinces is illustrated in Figure 3.1 below. Whereas summary traffic counts are
given in Table B.2 in Appendix B.

Carl Bro International a|s and Associates

10

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Figure C3.1: Distribution of Vehicle Types Observed in Three Provinces

45%
40%

% of All Vehicles

35%
30%
25%
20%
15%
10%
5%
0%
M/C

Car
W-Java

Angkot

Buses
C-Java

LGV
E-Java

MGV

HGV

Total

Source: JARNS Traffic Surveys

3.4.2

Traffic Count Data from Other Sources

Traffic data were also obtained from a number of other sources. Where the traffic count
location was common with the JARNS traffic surveys, the JARNS data was used. The data
for other locations was selected form different sources in the following order of preference.
1.
2.
3.
4.
5.
6.
7.

JARNS (2000)
Jasa Marga (2000)
ATC (1997 & 1998)
Pekalongan / Probolinggo SURIP Studies (2000)
TSSS (1998)
CAPEX (1997)
JARNS Estimates (2000)

In all cases, data was converted to the base year of 2000, using the average growth rates
experienced in each province. Also not in all cases, the available data was grouped into
classes of vehicles used by JARNS. These data were disaggregated to the same vehicle
categories as used in JARNS modelling work. A synthesis of all data sources was used to
update IRMS link traffic data. Traffic counts for all strategic links in the network are reported
in Technical Report 2.

3.5

Roadside Interview Surveys (RIS)

3.5.1

General Background

Roadside Interview surveys were conducted at 37 sites through out Java for the following
seven private vehicle types; listed below as four combined categories:

Private Car/ Jeep etc (Vehicle Type 2)

Light Goods Vehicles LGV (Utility Vehicle, Type 4)

Carl Bro International a|s and Associates

11

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Medium Goods Vehicles MGV (Goods Vehicle Types 7 & 8)

Heavy Goods Vehicles HGV (Goods Vehicle Types 9, 10 & 11)

The survey method entailed stopping a representative random sample of these vehicle types,
and interviewing the driver, regarding the origin and destination (O-D) of the trip. For private
vehicles trip purpose and number of vehicle occupants were also recorded. For goods
vehicles, commodity types and quantities carried were also recorded.
For Angkots and Buses the origin and destination (O-D) as specified by the route displayed at
the front of the vehicles was recorded. Observations of the number of occupants in the
vehicle were made, albeit in crude categories: empty, , or , or , or completely full. These
vehicle occupancy observations were then converted to average vehicle occupancy depending
upon the size and type of vehicle.
The O-D survey data recorded at sites was coded and entered into a computer database using
Microsoft EXCEL and Access software packages. Trip origin and destination addresses were
converted to JARNS study area zone system. All survey data was checked and validated
through a series of range, logic and consistency checks. In most cases, correction of coding/
data entry errors was possible. A small proportion of data found to be erroneous, was
rejected. Over all, the sample size achieved was statistically significant for the O-D data for
each vehicle category. The final sample size of the accepted data set, varied from site to sites,
and between categories of vehicles.
3.5.2

Car O-D Surveys

Table 3.3 below summarises the sample size after all the editing and validation of the
interview data. The overall sample size of 17% is considered very good. Even at the lower
end, in West Java the sample size was 9%, which is considered significant for such surveys.
Table 3.3: Car RIS O-D Survey Sample Size
Survey

West Java

Central Java

East Java

Total Java

RIS Final Sample

5,047

16,102

11,990

33,139

Total MCC Counts

56,806

86,277

53,768

196,851

9%

19%

22%

17%

% Sampled

Car occupancies were also recorded during the O-D surveys. Car occupancies varied from
site to site, and were observed to be as high as 12-persons per vehicle, with an average
occupancy of 2.7. Other attributes of car trips were also analysed, but were found to be not of
use to the modelling process.
3.5.3

Goods Vehicles O-D Surveys

The O-D surveys were conducted for all goods vehicles at the same sites as for Cars. The
sample size achieved is summarised below for the three aggregate goods vehicle categories.
The overall sample size achieved was considered adequate for this study. No particular
reason could be found why the sample sizes are low for sites in West Java. The final sample
sizes are summarised in Table 3.4 below.
The data on commodity types and quantities carried was also analysed. It was found that in
most cases it was not possible to make effective use of this data, as the drivers/conductors of
the vehicles were not willing to provide adequate/accurate information.

Carl Bro International a|s and Associates

12

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Table 3.4: Goods Vehicle O-D Survey Sample Sizes

RIS Final Sample

West
Java
1,723

Central
Java
8,823

East
Java
5,091

Total
Java
15,637

Total MCC Count

19,557

31,810

19,113

70,480

9%

28%

27%

22%

RIS Final Sample

3,576

9,016

7,453

20,045

Total MCC Count

36,617

46,682

27,727

111,026

% Sampled

10%

19%

27%

18%

RIS Final Sample

663

1,999

2,624

5,286

Total MCC Count

11,399

17,315

14,068

42,782

6%

12%

19%

12%

Goods Vehicle Category

Light Goods Vehicles
(Vehicle Type 4)

Survey

% Sampled
Medium Goods Vehicles
(Vehicle Types 7&8)

Heavy Goods Vehicles

(Vehicle Types 9,10&11)

% Sampled

3.5.4

Angkot O-D Surveys

Angkot O-D surveys were conducted by recording the origin and destination of the vehicles.
Analysis of the O-D data revealed that most of the Angkots travel over short distances
(average trip length was observed to be less than 40 km), normally between two or three
adjacent small towns. O-D data when converted to zone-to-zone movements showed that
high proportions of the trips are either intra-zonal (40%) or between adjacent zones. This
reflects that the passenger O-Ds were of even shorter distances; i.e. passengers get on and off
Angkots all along the route. Therefore, it was decided that it would not be appropriate to
model the O-D patterns of Angkot in terms of a matrix of movements, as there would not be
any route choice applicable to the travel of these vehicles. Instead, these vehicles were
modelled as fix loadings on the sections of the network where they operate. This is a better
way to reflect their use of the road space in the modelling process.
Vehicle occupancies of Angkots were also recorded. The analysis showed that some 15% of
these vehicles were empty. It was implied that these vehicles were travelling empty back to
get a new load of passengers. The Angkot occupancy varied between few passengers to
maximum of 15 passengers per vehicle over the survey sites. It should be noted that not all
Angkots are of the same design capacity. Some vehicles have seating capacity of more than
twelve passengers whereas other vehicles have seats just for 6-8 passengers. The occupancy
survey showed that close to 34% of vehicles were almost full carrying 12 or more passengers,
and about the same percentage carrying between 6 and 11 passengers. Average occupancy
was computed to be 7.6 Pax per vehicle. No noticeable differences in vehicle occupancy were
observed across survey sites or the regions.
3.5.5

Bus O-D surveys

Bus origin and destination surveys were conducted at all RIS locations. At four locations, it
was found that the number of buses is small therefore, no bus O-D were recorded. Separate
analyses of RIS data for the medium and large buses (vehicle types 5&6) were carried out.
The analysis showed that there are only a small number of medium size buses in operation on
inter-urban routes. It was also observed that, during the surveys the distinction between the
medium and the large bus was not clear at all times. Therefore, for further analysis and the
matrix building process the data for the medium and large buses were combined. In all
subsequent analyses, the category bus implies medium and large buses. The O-D sample
size achieved was high (60%). This is detailed for each province in Table 3.5 below.

Carl Bro International a|s and Associates

13

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Table 3.5: Bus O-D Survey Sample Size

Survey

West Java

Central Java

East Java

Total Java

Final O-D Sample

9,146

16,665

5,695

31,506

Total MCC Count

11,298

29,697

11,215

52,210

81%

56%

51%

60%

% Sampled

Bus occupancies were observed during the surveys by either counting the total number of
passengers in a bus or by recording the overall occupancy of the bus under one of the five
occupancy status, (Empty, , or , or , or full). The average occupancy was observed to be
33 Pax per bus, and was recorded to be as high as 70 Pax.

Carl Bro International a|s and Associates

14

Java Arterial Road Network Study (JARNS)

4.
4.1

Technical ReportNo.10:
Network Model Development and Application

DEVELOPMENT OF BASE YEAR VEHICLE TRIP MATRICES

Introduction

The base year trip matrices were developed from a number of data sources, and through a
complex series of processes. The starting point was the building of the observed trip matrices
from the RIS surveys. These observed matrices accounted for only a small proportion of the
trips that take place over the strategic network. The remainder of trips were, either imported
from other sources, or estimated using well-known Matrix Estimation techniques. This
Section briefly describes the complete process involving the development of the base year
trip matrices. In total five trip matrices are estimated, one for each of the five categories of
vehicles for a 24-hour period. Detailed tables and figures are also presented in Appendix C.

4.2

Development of Observed Trip Matrices

Roadside Interview Surveys recorded the origin-destination of vehicle trips as reported by the
drivers. Each cell in the trip matrix represents a combination of origin zone and destination
zone, i.e. O-D pair. The content of the cell is the number of trips for that O-D pair. The
observed matrix development process calculates the number of trips from the data collected
from the RIS survey stations. The series of steps that were followed in building the observed
O-D matrices are outlined below.
1.

Coding of Addresses: Code the location to a JARNS study area traffic zone; if it
was not possible to directly code the zone number due to incomplete address or
recorded information then enter the an area-code (i.e. Kecamatan / Kabupaten /
Kotamadya specific code). If an address or location could not be mapped, the
interview was rejected. Up to this stage due to incomplete information on O-D
addresses 2 to 3 percent of the interviews were rejected.

2.

Validation of Address Codes: Check if the coded O-D area code or zone number
is valid or not? If incorrect then go back to step-1, and check the O-D coding.
Repeat steps 1&2 until all O-D codes are acceptable, and then proceed to Step-3.

3.

Conversion to Zone System: Convert the interviews with an area code to the
zone numbers:
o
o

4.

Change area codes to zone number for areas such as Kecamatan/ or other
locations, which lie wholly within a zone;
Observed trips with O-D recorded as area code(s), that represented a group
of zones (such as a Kabupaten comprising of a number of zones) were
disaggregated to represent zonal trips in proportion to the population in each
zone.

Validation of an O-D Pair at A Survey Station: Check the validity of the O-D
zone pair, by checking if the trip passes through the survey station or not? If
O=D, i.e. the trip is intra-zonal then the O-D path will not exist. These trips were
included in the observed matrix. If the trip is inter-zonal (i.e. O<>D) and the OD path does not pass through the survey station, the observation was rejected.
These errors are quite common for such surveys. Because of the repeated editing
and updating the interview data, the number of interviews that were rejected was
quite low. The number of interviews that were rejected varied between survey
sites and was 5 to 10 percent of the sampled interviews.

Carl Bro International a|s and Associates

15

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

5.

Calculate Gross-up Factors: The accepted interviews represented only a sample

of total traffic passing a survey station. To represent all traffic, the sampled
interviews are grossed up to the total traffic. The gross-up factors were
calculated for each survey station, for each direction of travel, for specific time
bands over the survey duration. Usually the time band is in or one-hour
duration. In instances when the observed sample was too small, longer time
bands of up to 3-hour duration were used.

6.

Adjustment for Multiple Station Observations: A path between an O-D pair may
pass through more than one survey station, so it may be sampled at more than
one station. In principle, this was allowed for by allocating a multi-path factor to
each interview. The factor was computed to be equal to 1/(the number of survey
stations on a path).

7.

Adjustment Factor for Survey Period: At most locations the RIS were carried
out for a period of 12-hours. In order to obtain the full 24-hour trips, the 24hour/12-hour factor was calculated from the MCC surveys.

8.

Calculation of Total Observed Daily Trips: The total number of daily trips
observed between an O-D pair were then computed by multiplying the number
of valid interviews per O-D pair by the grossing-up factor, multiple station
observation factor, and the survey period factor.

The observed trip totals for each vehicle category are summarised in Table 4.1. Car trips
accounted for 43% of all observed trips. The second largest category was medium goods
vehicles. All goods vehicles accounted for 46% of the observed total trips, whereas the bus
volumes on inter-urban roads were low and accounted for only 11% of the observed trips.
Intra-zonal trips were observed at a few survey sites that were located inside the zone
boundary due to safety reasons. These trips account for 5% of the total observed volumes.
The proportion of intra-zonal trips is low for vehicle categories that make shorter journey
than those which travel over longer distances, such as large buses or heavy goods vehicles.
Table 4.1: Summary of 2000 Observed Total Trips by Vehicle Category (AADT)
Vehicle
Category

Total InterZonal Trips

Observed

Total IntraZonal Trips

Observed

% IntraZonal

Total
Trips

% of
Total
Trips

Car

166,023

9,407

5%

175,430

43%

LGV

60,399

4,511

7%

64,910

16%

MGV

87,736

4,716

5%

92,452

23%

HGV

28,891

462

2%

29,353

7%

Bus

42,443

896

2%

43,339

11%

385,492

19,992

5%

405,484

100%

All Vehicles

The total trips represented only a fraction of the zone-to-zone (264x264) O-D pairs, which
the matrix represents. In order to intercept all the O-D pairs, a large number of surveys would
be required and this was beyond the resources of the study. However, these observed O-D
pairs do provide a sound basis to estimate the trips between the remaining unobserved O-D
pairs. This process is described next.

Carl Bro International a|s and Associates

16

Java Arterial Road Network Study (JARNS)

4.3

Technical ReportNo.10:
Network Model Development and Application

Synthesis of Base Year Trip Matrices

There are a number of analytical techniques, which could be adopted for the synthesis of the
unobserved O-D pairs. Some of these techniques rely wholly on the observed matrix and the
associated network, but require an estimate of the total number of trips in the whole of the
study area, which is usually difficult to provide in the absence of any household or other
survey data. Other techniques rely on an initial matrix, and a series of traffic counts at key
screenlines and cordons in the study area. From the outset, the JARNS study approach was
based on the later technique of estimating the trip matrices from the road traffic interview and
count surveys. The full matrix estimation process as developed for the JARNS study was
based on a series of well thought-out process, which made extensive use of not only the
traffic survey data but also of O-D and traffic data available from other studies and databases.
The JARNS matrix estimation methodology comprised three main stages. The same
methodology was adopted for the estimation trip matrices for all five categories of vehicles.
4.3.1

Build Prior Trip Matrices

This stage involved updating the five (Car/ LGV/ MGV/ HGV & Bus) observed AADT trip
matrices to include trips for as many O-D pairs as possible from a number of available
sources. This required that the trips to be included into the JARNS observed trips matrices
must be compatible in terms vehicle category and the zone system. The data sources available
were:
1.
2.
3.

JARNS Observed O-D trip matrices,

TSSS 1998 total vehicle Inter-Kabupaten trip matrix,
TSSS 1998 Inter-provincial matrix of ferry traffic (for external trips to/from Bali
and Sumatra)
4.
SURIP studies trip matrices, and
5.
1996 NODS includes three basic trip matrices of annual inter-Kabupaten travel:
o Total private road traffic O-D, vehicles and passengers (excluding
motorcycles);
o Total public road traffic (i.e. buses and Angkots) O-D, vehicles and
passengers; and
o Total goods traffic O-D vehicles and tonnage.
The observed O-Ds were given the highest level of confidence, and all trips were included in
the prior matrix. The data from other sources was reviewed and analysed. The conclusion
was that only the TSSS data is of suitable standard and level of detail, and could be
incorporated into the prior trip matrices, albeit after some manipulations. The data from
SURIP studies was either too detailed or the O-D pairs had already been observed by in the
JARNS survey. Therefore, no O-D trip data from SURIP studies was included. It is clear
from a number sources and analyses that the 1996 NODS data is not reliable, and to some
extent is outdated. Therefore, no further use of the 1996 NODS trip matrices was made.
The key steps involved in building the prior matrices are outlined below.
1.
2.
3.

Build symmetrical RIS matrices for each vehicle category, and all vehicles
combined.
Sector RIS matrices (22 Sectors) and calculate percentage of trips by Car, LGV,
MGV, HGV and Bus as percentage of the total trips for each sector-pair.
Convert TSSS 1998 all vehicle trip matrix to 2000, by applying an average
growth factor of 1.088, and expand the inter-Kabupaten trips to JARNS 264
zone system. The expansion factors were based on the distribution/ proportion of
population in each zone within the Kabupaten.

Carl Bro International a|s and Associates

17

Java Arterial Road Network Study (JARNS)

4.
5.

6.

7.

Technical ReportNo.10:
Network Model Development and Application

Disaggregate the TSSS total vehicle trip matrix into five vehicle categories,
based on sector-to-sector percentages derived in Step-2.
RIS captured a reasonable number of Java-Bali trips; but a few Sumatra- Java
trips. Vehicle trips for each category of vehicle, from both external zones (Bali
& Sumatra) were estimated from the TSSS ferry trip matrices and disaggregated to zones using proportions based on population in each zone.
Prior matrix was then formed by selecting all trips from RIS matrix, and
including trips from TSSS matrix if an O-D pair was not observed. Similarly,
unobserved O-D pairs between internal and external zones were added from the
dis-aggregated TSSS ferry traffic matrices.
O-D pairs with zero values were given a seed value of 0.1 trips.

The Table 4.2 compares the trip totals in RIS, TSSS and prior matrices for each of the five
vehicle categories. The increase in trips from RIS to the prior matrix indicates how the
observed matrices missed more that half of the total trips, which were taken from the TSSS
trip matrices.
Table 4.2: Summary of RIS, TSS and Prior Matrices Trip Totals
Vehicle
Category

RIS
Total Trips

TSSS
Total Trips

Prior Mx
Trip Totals

%
Increase
from RIS

Car

175,430

370,297

397,030

126%

LGV

64,910

106,609

128,591

98%

MGV

92,452

183,258

207,110

124%

HGV

29,353

45,887

60,740

107%

Bus

43,339

66,183

91,638

111%

405,484

772,234

885,109

118%

All Vehicles

4.3.2

Initial Matrix Estimation

Matrix Estimation takes several inputs, which provide information on number of observed/
existing trips, trip patterns, traffic counts, a set of paths that define the likelihood of an O-D
pair passing through count site(s). It also takes user-defined confidence levels associated with
each element of the input data. The estimation process then computes a new trip matrix that
fits the input count data best. The user can change the input confidence levels, to control the
output. A number of criteria are employed to check the output results, and declare that the
output matrix is the best estimate that could be achieved. These involve checking and
comparing the observed, and output trip length frequency distributions, comparison of
observed and assigned traffic volumes at input count stations and at locations not included in
the estimation process, and check of reasonableness on certain known area-to-area
movements.
Matrix Estimation was carried out in two parts. The inputs to the initial matrix estimation
process for each vehicle category were:

Prior trip matrix, as reported above;

Trip ends taken as the average of the sum of row and column totals of the prior
matrix;
Trip ends seeded with 50 trips if no observation, (10 for HGV)
A set of five Generalised Cost (GC) paths built using a Burrell technique with
a spread of 100. The GC values used are summarised in Table 4.3 below.

Carl Bro International a|s and Associates

18

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Table 4.3: GC Parameters for Building Path for Matrix Estimation

Vehicle
Category

Value of Time
(Rp/Hour)

Vehicle Operating
Cost (Rp/Km)

Average Toll
Payment
(Rp/km)

Car

22,300

80

100

LGV

9,300

120

100

MGV

45,000

110

100

HGV

53,000

180

200

Bus

120,000

100

160

Source: Consultants Estimate, all values in 2000 prices

Traffic count data for a number of links in the network was input to the matrix
estimation process. It was input for 16 screenlines defined to capture east-west
movements and for 10 cordons around major urban areas (see Figure 5.9).
Screenlines and cordons were designed to include as many links as possible with
JARNS traffic counts, and avoid links for which count data was taken from
other sources. The traffic counts were input to the matrix estimation as counts on
individual links (not as aggregates for screenlines/ cordons) as in most cases
parallel links served different pairs of O-D movements.

This initial matrix estimation stage yielded a trip matrix that provided a good estimate of
inter-Kabupaten / inter-zonal trips. The process could not estimate intra-Kabupaten / interzonal trips because the prior matrix contained few observations for such O-D pairs. The
TSSS matrix had no intra-Kabupaten trips, and the RIS matrix had a small number of intrakabupaten trips.
4.3.3

Final Matrix Estimation

Subsequent to the first stage estimation intra-Kabupaten / inter-zonal trips were extracted
from the estimated matrices and tabulated vs. population and employment for the
corresponding Kabupaten. Models were derived relating intra-Kabupaten / inter-zonal trips
to population and employment. Regression models were examined but coefficients were not
significant and R2 values were too low. Therefore, average trip rates vs. employment
(population for bus) were calculated. The estimated matrices from the previous stage were
in-filled with estimated intra-Kabupaten / inter-zonal trips. These in-filled matrices were now
input with the same traffic counts and the other data as specified above to this second/ final
stage of matrix estimation process. The process was repeated for each vehicle category
separately.
The output estimated matrices were then assigned to the network. The assigned volumes were
compared against the traffic counts. Upon achieving a satisfactory comparison, the matrix
estimation process was terminated. If the comparison of assigned volumes were significantly
different, the whole process was repeated, with amended input data. The amendments
amounted to either updating the network or changing the confidence levels associated with
the input data, input to the estimation process.
The Table 4.4 below compares the final estimated trip totals with the RIS, TSSS, prior and
initially estimated trip totals. It can be seen that the number of trips did not change by much
between the two stages of the estimations. However, the final matrices did provide a much
better comparison of the observed and assigned traffic volumes on the network. These results
are reported in detail for each vehicle category in Technical Report 10.

Carl Bro International a|s and Associates

19

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Table 4.4: RIS, TSSS, Prior and Estimated Matrices Trip Totals for 2000 (AADT)
Vehicle
Category

RIS
Total
Trips

TSSS
Total
Trips

Car

175,430

370,297

LGV

64,910

MGV

Prior Mx
Trip
Totals

Initial Estimation
Trip Totals

Final Estimation
Trip Totals

397,030

354,233

350,448

106,609

128,591

109,360

114,363

92,452

183,258

207,110

151,901

157,396

HGV

29,353

45,887

60,740

35,451

38,208

Bus

43,339

66,183

91,638

44,623

45,457

405,484

772,234

885,109

695,568

705,872

All Vehicles

It may be of some concern that the TSSS all-vehicle trips total is about 9% more than the
JARNS 2000 final estimated trips. This was investigated further. The main reasons were
found to be:

The TSSS estimated a single matrix of all vehicles based on the 1998 counts for
some 75 sites grouped into 19 screenlines. The JARNS matrix estimation is
based on over 150 count sites.
The TSSS estimate was for much coarser zone system of 98 zones, whereas as
JARNS study area has a much more refined system comprising of 264 zones.
The TSSS used a much simplified highway network, treating many parallel
roads as corridors. Whereas the JARNS highway network is more detailed,
covering a much wider selection of route choices, and is based on a more
detailed database.
In any case, the difference of 9% is well within the margin of error of data used
for such strategic studies.

These final estimated trip matrices were used to form the basis of JARNS network model
validation, and the development of forecast models. All trip matrices are summarised in
Appendix C.

Carl Bro International a|s and Associates

20

Java Arterial Road Network Study (JARNS)

5.

Technical ReportNo.10:
Network Model Development and Application

JARNS NETWORK MODEL DEVELOPMENT AND VALIDATION

5.1

Introduction

A key component of the modelling process is the development of a road network model that
would allow the forecasting of future traffic flows on the strategic network. The network
model should be detailed enough so that it can adequately represent the strategic inter-urban
traffic in Java. It should also be robust enough to test a wide range of different network
development strategies. An accurate representation of the road network and its attributes is
therefore essential. It is the network model from which travel time, cost, distance are taken
for the development of the trip matrices, and are essential inputs to the economic evaluation
framework for testing alternative network strategies/ scenarios.
The scope of the study area network has been outline in section 2 of this Appendix and it is
illustrated in Figure 5.1 by type of roads. This sections elaborates on the development of the
road network and its elements, and its use in the over modelling process.

5.2
5.2.1

Network Data Sources and Coding

Network Model Data Sources

The development of the study area network model relied heavily on the IRMS and ARMS
databases. Data on a number of road characteristics were also obtained from the ARMS
database of digital images of roads. Figure 5.2 shows the flow of data, the interaction and
interfaces between the IRMS/ARMS and the TRIPS network model. How different elements
of the network were developed from different data sources and combined to form a complete
network model are described in the next section. Further details are given in Appendix D.
5.2.2

Coding of Study Area Network General

The coding of the network model involves representation of the selected roads network into
a form acceptable to the modelling process. A highway (road) network in the TRIPS model is
represented by links (representing sections of roads) and nodes (represents junctions or the
end of a road section). In a network model, every link must be connected to the adjacent links
at both of its ends to form a continuous system of links called a network.

Carl Bro International a|s and Associates

21

Merak(

JAKARTA

Serang(

>

)
)

Cikampek
)

)
)

)
)

Bogor.

)
)

Sukabumi.

BANDUNG.

Tegal
)

Rembang.

Cirebon.

Pekalongan
(

SEMARANG.

Tuban

)
)

Sumenep.

)
)
)

Salatiga.

Ciamis.

Magelang.

Cilacap.

Jombang

Surakarta

n
SURABAYA

YOGYAKARTA

)
)

)
)

( Gempol
)

Madiun

Probolinggo

)
)

)
)

(Malang.

Tulungagung

)
)

)
)

Banyuwangi.

Pacitan

LEGEND :
Single-2 < 6.7 m
Single-2
Single-4
Dual-2
Dual-3
Toll Dual-2
Toll Dual-3
Toll Dual-4
Toll Access Links
Urban/Kota Links
Urban Corridors
Rural ( Non-Strategic Roads )

Province Boundary

FIGURE 5.1: BASE YEAR (2000) NETWORK

BY ROAD TYPES

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Figure 5.2: IRMS/ARMS Interface with TRIPS Network Model

TRIPS

Exogenous Estim ates

from IHCM ,
IRM S/ARM S

Travel D em and
M atrices

IH CM

A verage
PCE
Factors

Travle D em and
M atrix of Trips
in PCU /hour

Road Type
Terrain
Pavem ent &
Shoulder W idths
Side Friction

Link
Capacity

A ssignm ent
V olum e/
Capacity

Road Type
Terrain
Pavem ent &
Shoulder W idths
Side Friction
Roadside
Land U se

Free Flow Speed,

and Speed Flow
Curve No.

Congested
Speeds

The only links that do not have both ends connected to other links are notional links called
Centroid Connectors. These links are used to load traffic onto the network from the zone
system, and the open end of these links is represented by that zone. A link in the TRIPS
network model should be roughly homogeneous in terms of its physical and operational
characteristics along its length. A contiguous road section (may be without junctions) with
variable characteristics along its length may be split into two or more links in order to
adequately represent the features that influence the flow of traffic. Each link in the TRIPS
model requires the coding of the following fields:
1.
2.
3.
4.

Nodes: Each link is represented by two nodes, usually called as Anode and
Bnode. All nodes are numbered using a 5-digit code representing its location in
the study area.
Distance: Represents the link length in 10m units between Anode and Bnode.
Link type: A 2-digit code representing the road type, used for tabulations.
Jurisdiction Code: A 2-digit code that may be used to represent different aspects
a link, e.g. its location. In JARNS model, it is used to represent road
classification specifically used for the calculation of accidents for the evaluation
framework.

Carl Bro International a|s and Associates

22

Java Arterial Road Network Study (JARNS)

5.
6.
7.

Technical ReportNo.10:
Network Model Development and Application

Capacity Index: A 2-digit code used to represent the number of the Speed/Flow
curve for that link. Details in Section C5.3 on Link Capacities and Speed/Flow
Relationships.
Link Speed: Base year link speed, computed from the base year traffic volume
and the speed flow relationship.
Link Capacity: 1-way link capacity. Details in the Section C5.3 on Link
Capacities and Speed/Flow Relationships.

The coding of these seven items for over 900 links of the JARNS network model required
considerable planning and data manipulation. The sources of data for each of these seven
items differ, depending upon the level of detail and function/importance of these links in the
network model. The coding of these values is described in the following sub-sections.
5.2.3

Coding of Strategic Network Links

Network coding requires representing the features of each link in the network. The level of
detail required to code these features depends upon the function and importance of these links
in the overall network model. For this study, the most important part of the network coding
was the adequately detailed and accurate representation of the roads defined as strategic.
The JARNS strategic network covered 7372 km inter-urban roads and includes 406 IRMS
links.
The majority of the junctions between arterial and collector roads are in cities. The IRMS
database defines the sections of road between the formal administrative boundary of large
city and the junction with another road within the city as urban links. This is illustrated below
in Figure 5.3. The coding of these IRMS urban links is defined in a later section. This section
deals with the coding of the inter-urban links.

Carl Bro International a|s and Associates

23

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Figure 5.3: IRMS Inter-urban and Urban Link Illustration

IRM S Urban Links

IRM S Inter-urban Links

22022

Formal Administrative
Boundary of Urban Area

220241
2202511
22013
2202512

22067

It was possible to code a number of these inter-urban links directly in to the TRIPS network
However, in instances of small towns/ urban areas without formal administrative boundary,
this division had not been carried out and the IRMS links continues right through the
urban/developed areas. For the modelling of these roads, it was necessary to reflect the
changes in the physical and traffic characteristics that occur on the sections within these
urban/developed areas. This is further illustrated in Figure 5.4 below. Therefore, the sections
of the IRMS links which continue into the urban/ developed areas (shaded) would have
different characteristics, and need to be represented separately.

Carl Bro International a|s and Associates

24

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Figure 5.4: Illustration of IRMS Inter-urban Link Subdivision

IR M S Inter-urban Links

U rban A rea W ithout Form al A dm inistrative Boundary

Subdivide IR M S Inter-urban L inks

IR M S Inter-urban Links
TR IP S Inter-urban Links
TR IP S U rban Links

This subdivision of these links and their representation was a major exercise, and required
detailed information about the start/ finish points of the urban area. This task was carried out
using the ARMS digital image data. This data provide a digital image of all road links at
every 10m intervals. The data was scanned and information on roadside landuse (in %), side
friction, and road type was recorded for every 100m section, or wherever there was any other
reason to subdivide the IRMS link. For each of the 406 IRMS links data was recorded on
forms. The data was then collated, to subdivide the IRMS links into sections equivalent to
TRIPS links at the boundaries of towns/ developed and undeveloped areas. The data was
coded for each section of the IRMS link. The Table below summarises the data coded for
each of the strategic TRIPS link.
The JARNS network retained the unique IRMS Link-ID code with additional section
numbers through out its own network database. This would maintain the much needed
compatibility between the JARNS and the main IRMS/ARMS databases. The data items
extracted from the IRMS/ARMS databases, and their use in the JARNS network model is
given in Table 5.1.

Carl Bro International a|s and Associates

25

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Table 5.1: Link Attributes Taken from the IRMS and ARMS Databases
IRMS/ARMS
Database Field

JARNS
Database

TRIPS
Model Field

Used for
Speed Flow
Relationship

Link ID

Link-ID

No

Section

No

Province Code

Prov-ID

Jurisdiction
Code

No

Link Length
(in 1m units)

Length

Distance

No

ROAD_TYPE

Link-Type

Link Type

Yes

FUNCT

Function

TERRAIN

Terrain

Yes

WCARR

Carr-Width

Yes

Shld-Width

Yes

LandUse

Yes

Rd-Friction

Yes

Link-Type

Link Type

Yes

Int-Dev

SH_WIDTHL &
SH-WIDTHR
ARMS Digital Images
(Roadside Landuse)
ARMS Digital Images
(Side Friction)
ARMS Digital Images
(Road Type)
ARMS Digital Images

Comments
A code created by concatenating
the Province, Ruas and Suffix
Numbers
A 2-digit section number, where
IRMS/ARMS Links has been
Subdivided.
22/24/26/28
Rounded and Coded in
10m units
2/3/4/5 for undivided
IRMS Links
A, K1, K2, K3
(Source IRMS)
Flat, Rolling, or Hilly
2-Way Width in
Decimetre (xx.x m)
Averaged of both Sides
Decimetre (xx.x m)
Roadside Landuse (%)
Side Friction
(1-very low, to 5 very high)
2/3/4/5 as above for
Sections of IRMS Links
Code I for Inter-urban, or
D for Developed Sections

The main items coded directly into the JARNS database and TRIPS network model were:

Distance,
Road Type, and
Jurisdiction code, road classification based on road type and carriageway width
used for accident calculations.

The other data items were used to calculate the following attributes, and used in the JARNS
database and TRIPS network model. These were:

Road Capacity
Free Flow Speed,
Capacity Index based on the Free Flow speed, and
Link speed based on the base year capacity traffic volumes.

This effectively was the complete coding of the strategic network. In total, the network
comprised of 808 TRIPS model links, representing the 406 IRMS links covering 7372 km of
strategic road network of Java.
5.2.4

Coding of Urban Area Links

These are sections of national road network, which extend into the urban areas (see Figure
5.3). These links are treated by the IRMS/ARMS database as urban/ Kota links, and are

Carl Bro International a|s and Associates

26

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

classified as such. These links are identified in the IRMS database by the use of special 2digit suffix and by a character, K in the special filed called Kota in the IRMS database.
These links form an integral part of the JARNS/ TRIPS model network. Their function in the
network model is to facilitate the movement of inter-urban traffic into the urban areas, and
allow the through traffic to pass. These links carry both the local and strategic (i.e. Interurban) traffic. It was not within the scope of this study to fully model the local traffic.
However, the account of the local traffic was taken by coding these links under a special
category, with reduced capacity equal to 50% of a divided 4-lane urban road, and base year
speed of 30 km/h. This process was validated by analysing the data of SURIP studies, which
have modelled both the local and through (inter-urban) traffic explicitly. Under this category
of Urban area roads, 123 IRMS links representing close to 414 km of urban roads of Java
were represented in the network model by 142 TRIPS links.
5.2.5

Coding of Urban Area Corridors

The network coding described so far covered majority of the inter-urban network and
substantial part of the urban links. However, it did not include links through the major
metropolitan areas of Jakarta, Bandung and Semarang, because the urban streets of these
cities are not included in the IRMS database. The major urban area roads of these cities are
represented by notional links in the TRIPS network model, and are called urban area
corridors. These corridors provided the same function i.e. passage of inter-urban traffic into
and through these cities, as did the urban links. The coding (Capacity, Speed and Speed/
Flow relationships) of these links was therefore similar to that of the urban area links as
described above in Section C5.2.4. Urban corridors incorporated in the network had a total
length of 182 km.
5.2.6

Toll Roads and Toll Road Access Links

All urban and inter-urban toll roads of Java are included in the JARNS network. Wherever
the toll roads have intersection with the strategic network, the access/ egress to the toll roads
is via toll access links. The toll access links are also notional links of 0.5 km length, or longer
depending upon the distance of the toll road from the nearest strategic inter-urban link. The
inter-urban toll roads were coded with full details, including number of lanes etc. The out of
pocket cost of toll charges paid by the users are also model fully by representing the toll
roads under special category link types. The coding of toll access is also similar to the coding
of the toll roads, but these links do not have the same capacity as the toll road. The JARNS
base year model network included 439 km of toll roads and 67 km of access toll road access
links.
5.2.7

Rural Area Roads

JARNS network model also included 602 km of rural (non-strategic) roads from the southern
part of West Java province. These roads are not treated as strategic roads, but do perform
vital function of providing accessibility to these remote areas. These links represented 572
km of K2, and 30 km of K3 road.
5.2.8

Summary

The complete representation of the JARNS model network is given in Table 5.2 below, and it
is illustrated in Figure 5.1 representing each category of roads in the network.

Carl Bro International a|s and Associates

27

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Table 5.2: JARNS Model Network Description by Province and Length of Roads (km)
Road Function in the JARNS
Model Network
Inter-urban
Sections
Urban-area
Sections
Sub-total Strategic Links

Strategic
Network
Links

Cent.
Java &
Yogya.

West
Java

Jakarta

East Java

Total
Network

1720

2,532

2,052

6,304

344

387

337

1,068

2,064

2,919

2,389

7,372

IRMS Urban

83

135

122

340

IRMS Kota

22

52

74

Toll Roads

88

269

20

62

439

Access Links

53

67

Urban Area Corridor Links

153

27

182

Rural Roads & Ferry Link

602

609

248

3,098

3,100

2,637

9,083

Urban Area
Links
Toll Roads

Total Network

5.3

Link Capacity and Speed Flow Relationships

The calculation of link capacity and the application of speed/flow relationships are crucial in
any network modelling, as these relationships dictate the vital statistic relating to network
costs and its operation. These statistics are also important for the economic and operational
evaluation of network upgrade scenarios. Link capacities and speeds were available from the
IRMS/ARMS database. A detailed review showed that these values are outdated, and need to
be revised. The IRMS values were based on the IHCM relationships. For the JARNS network
model, these relationships were reviewed and simplified to suit the modelling requirements,
and the available data. The link-based attributes used for the calculation of capacity and
speed/flow relationships are identified in Table 5.1 above, and a brief description is given
below, and full details are given in Appendix E.
Link Capacity Function of (Road-type, Carriageway and Shoulder widths, Terrain Type,
and Side-friction)
Link Free Flow Speed Function of (Road-type, Carriageway and Shoulder widths, Terrain
Type, Side-friction, and Adjacent Land-use)
Inspection of the computed Free-Flow speeds of inter-urban links shows the following
ranges:

2 Lane roads: 30 75 km/h

Multi-lane roads: 50 90 km/h

Using increments of 5 km/h, speed/flow relationships were derived using the IHCM model of
three points per curve, these are:
1.
2.
3.

Volume/Capacity = 0.0; Speed = Free Flow speed;

Volume/Capacity = 0.85; Speed = As per IHCM for 2 or Multi-lane roads;
Minimum speed set to 10 km/h; and V/C at 10 km/h was calculated from the
relationship given in the IHCM.

In total 19 Seed/Flow, relationships covering all road types were defined. Figure 5.5 and
Figure 5.6 show these Speed/Flow relationships for 2-lane, and multi-lane roads respectively.

Carl Bro International a|s and Associates

28

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Figure 5.5: Speed Flow Curves for 2-Lane Roads

80
70

Speed (km/h)

60
50
40
30
20
10
0
0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.3

1.4

1.5

1.5

Volume Capacity Ratio

Figure 5.6: Speed Flow Curves for Multi-Lane Roads

90
80

Speed (km/h)

70
60
50
40
30
20
10
0
0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.6

Volume Capacity Ratio

5.4

Base Year Assignment and Model Validation

This stage of the modelling process brings together the travel demand (specified in terms of a
trip matrices) and the network. The assignment process then determines the route choices
people make on the network, and how the choice of route changes in a congested network
environment. The JARNS model used the well-known technique called Equilibrium
assignment. The key feature of this method is that it is a network cost optimisation process,
and it continues to iterate until there is no route of lower cost of travel between an O-D pair.
This technique is available as a standard program in the TRIPS package, and was adopted for
the JARNS network assignment model.

Carl Bro International a|s and Associates

29

Java Arterial Road Network Study (JARNS)

5.4.1

Technical ReportNo.10:
Network Model Development and Application

Assignment Time Period and PCE factors

The assignment process is carried out for a one-hour period for all vehicle types combined, as
the link capacities are usually specified on a per-hour basis in a common unit for total traffic
(i.e. Passenger Car Units PCUs). The (Passenger Car Equivalent) factors used to convert the
vehicle trips to the common unit of PCUs, were taken from the IHCM. The factors adopted
are reported below in Table 5.3.
Table 5.3: Average Hourly and PCE Factors, and Assigned Traffic Volumes

Motorcycles

24-Hours
Inter-Zonal
Trips
(AADT)
Observed Counts

Angkots

Observed Counts

Vehicle
Category

Average
Hourly
Factor (%)

PCE
Factor

6.57

0.5

Average
Hourly PCUs
Inter-Zonal
Assigned
Link Pre-loads

5.85

24-Hours
Total Trips
(AADT)

1.0

Link Pre-loads

Car

350,448

342,668

5.62

1.0

19,258

LGV

114,363

110,921

6.01

1.0

6,666

MGV

157,396

153,739

5.02

1.65*

12,734

HGV

38,208

36,774

4.30

3.8

6,004

Buses

45,457

44,781

4.85

1.67*

3,628

705,872

688,853

5.38*

1.29*

48,290

All Vehicles

* Weighted Average Values

The JARNS traffic count data was analysed to determine what time period should be
modelled. Figure 5.7 below shows the changes in total traffic volumes on the inter-urban
roads by time of day. It is evident that apart from some minor peaks in the traffic volumes in
the morning (06:00-08:00) and to a lesser extent in the late afternoon period the overall interurban traffic volumes are very similar through out the day. Therefore, it was decided to
assign the average hourly traffic volumes, rather than a particular peak period. However, the
proportions of vehicle types in the traffic stream during the day were observed to have some
variations. The proportions of vehicles in the average hour of the day were therefore
estimated separately for each vehicle category from the survey data. These average hourly
proportions are given in Table 5.3 above.

Carl Bro International a|s and Associates

30

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Figure 5.7: Distribution of Vehicle Volumes During the Day at 100 Survey Sites
6.5%

% in 1-Hour

6.0%
5.5%
5.0%
4.5%
4.0%

10

11

12

13

14

15

16

17

Time of Day (Hours)

Tot-PCU

5.4.2

All Vehs

Assignment Results Validation

The assignment of five vehicle categories combined was carried out by assigning the total
average hourly PCU trip matrix using the equilibrium assignment technique. The route choice
parameters were based on Generalised Cost (GC) of travel incorporating the value of time,
vehicle operating cost and the toll charges. These parameters were estimated for all vehicles
combined and weighted by vehicle mix. The values used in the assignment process are per
PCU in Rp, at year 2000prices, and are listed below:

Value of Time (VOT) = Rp 450/minute, (Rp 27,000/hr)

Vehicle Operating Cost (VOC) = Rp 70/km, and
Toll Charges = Rp 100/km of travel on the Toll roads and Toll Access Links.

The equilibrium assignment process converged after 10 iterations. The assignment results
were then compared with the observed traffic volumes. The assigned traffic volumes on the
inter-urban toll road links were found to be consistently below the observed volumes. This
aspect of under-assigning of traffic to toll roads is quite common when using a single route
choice GC for all road users. In order to improve the assignment results, different values of
VOT for the toll road users and other links were tested. It was found that the assignment
when the VOT for toll road users is at 90% of the value used for other roads gives the best
results. Therefore, for the base year and for the subsequent future year testing the VOT for
toll road users was taken as 90% of the VOT for the use of other roads.
The final base year assignment results are presented in detail in Appendix F. Table 5.4
compares the assigned and observed traffic volumes for a set of cordons and screenlines. This
comparison is also illustrated graphically in Figure 5.8. Both these comparisons show that the
assignment results show a close comparison between the observed and the assigned traffic
volumes. This shows that the modelling methodology is robust enough to be used for future
year testing of network scenarios.

Carl Bro International a|s and Associates

31

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Figure C5.8 Comaprison of Observed and Assigned Traffic Volumes (PCU/hr)

Assigned Volumes (PCU/hr)

12,000
10,000
8,000
6,000
4,000
2,000
-

2,000

4,000

6,000

8,000

10,000

12,000

Observed Counts (PCU/hr)

Table 5.4: Comparison of Observed and Assigned Traffic Volumes (PCU/hr)

Traffic Volumes (PCU/hr)
Difference
Percent
Screenline/
Cordon*
Observed
Assigned
(A - O)
% (A/O)
1
2,837
2,760
-77
2
6,823
6,935
112
3
4,314
4,615
301
4
2,967
3,116
149
5
2,786
2,939
153
7
1,984
2,108
124
8
2,913
2,907
-6
9
2,463
2,717
254
10
5,364
5,404
40
11
2,420
3,030
610
12
2,256
2,512
256
13
3,695
3,852
157
14
5,114
5,022
-92
15
1,926
1,856
-70
16
1,118
1,251
133
17
9,989
10,127
138
18
3,872
4,043
171
19
5,424
6,464
1,040
20
3,027
3,411
384
21
6,282
6,246
-36
22
4,015
4,018
3
23
4,134
4,274
140
24
7,324
7,430
106
25
7,191
7,064
-127
26
3,102
3,092
-10
Sumatra Ferry
466
486
20
Madura Ferry
296
336
40
Bali Ferry
262
248
-14
NB: Traffic Volumes Exclude Motorcycles and Angkots, which are Treated as Pre-Loads;
* See Figure 5.9 for Location of Cordons and Screenlines

Carl Bro International a|s and Associates

-3%
2%
7%
5%
5%
6%
0%
10%
1%
25%
11%
4%
-2%
-4%
12%
1%
4%
19%
13%
-1%
0%
3%
1%
-2%
0%
4%
14%
-5%

32

2
3
4

5
6

17
19
18

20

11

10

12

13

21

23
22

14

24

15
25

16

26

LEGEND :
Screenline
Cordon
JARNS Model Network
JARNS Model Network

FIGURE 5.9: NETWORK MODEL CORDONS AND SCREENLINES

Java Arterial Road Network Study (JARNS)

6.
6.1

Technical ReportNo.10:
Network Model Development and Application

FUTURE TRAVEL DEMAND

Introduction

The demand for inter-urban travel by road is a function of a number of variables related to
both the users and the provision of the infrastructure. The analyses of the base year road
travel demand and the studies of demand by other modes of travel (Railways and Coastal
Shipping; see Appendix E and Technical Reports 5 &6) showed that:

The demand for inter-urban travel is generally for business purposes.

It is usually for travel between adjacent cities.
There is limited long distance travel by road, and
The competition from other modes is not significant.

Therefore, the development of the forecast models was kept as simple as possible. This
simplicity was further dictated by the limited availability of good quality data, describing the
characteristics of the trip maker (i.e. the socio-economic data related to the traveller) and the
characteristics of the trips made.
The demand forecasting methodology used a three-stage process:
1.
2.
3.

The first stage estimates the growth in travel demand for each traffic zone in the
study area from the Trip End Models, which relate the demand for travel to the
independent socio-economic variables related to the trip makers in a zone.
The second stage distributes these trip ends over the study area, and yields the
forecast year trip matrices; and
The third stage then assigns these future trip matrices to the future test networks
to give the traffic volumes on these networks.

The major input to this model development process was the socio-economic data summarised
in Appendix A, and presented in Technical Report No.9.

6.2

Trip End Models

A number of different methods may be used for the development of Trip End model. These
differ according to the level of detail, availability and quality of data. The analysis of the base
year data showed that it would not be possible to develop trip end models in the conventional
way for both (generation and attraction) ends of a trip, because observed data lacked the level
of detail, and explanatory variable data was too coarse. The development of the trip end
models was for symmetric trip ends (i.e. Origins=Destinations; or Oi=Di;). These models
were developed for each of the five vehicle categories.
The methodology adopted was based on stepwise multiple-linear regression. Initially the
independent variables showing the highest correlation with the trip ends were identified.
These variables were then input to the step-wise multiple-linear regression approach. The
best-fit final model took the form of:
TEi = a * V1b * V2c * V3d
where: TEi : is the Number of Trip ends in Zone i
V1, V2 & V3 are independent variables, and
a, b, c, & d are parameters to be calibrated.

Carl Bro International a|s and Associates

33

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

It was found that the zonal population, urban-population and the zonal GDP are the best
explanatory variables. Initially attempts to develop Java-wide models when applied did not
produce acceptable results. This was found to be due to regional variations and disparities in
GDP and population densities. Separate sets of models were developed for West Java and
Jakarta, Central Java and Yogyakarta and East Java. The application of these models to the
base year planning data showed that they consistently under predict inter-urban demand for
large cities and somewhat over estimate demand for remote zones. A comparison of the
observed and modelled trips ends showed that these models cannot be directly used to
estimate absolute levels of demand, but would be appropriate to estimate the change in
demand due to socio-economic changes in a zone. The method used to estimate the future
year zonal trip ends then took the form:
TEFY = TEBY * (S_TEFY / S_TEBY)
where: TEFY are Future Year Trip Ends for Zone i
TEBY are the Base Year Trip Ends in Zone i
S_TEFY are the Synthesised Future Year Trip Ends for Zone i
S_TEBY are the Synthesised Base Year Trip Ends for Zone i
This process yielded the zonal trip ends that reflected the effects of estimated changes in
socio-economic data between the base and the forecast years. This overall growth in traffic
was estimated by examining the change in total trip ends for each vehicle category. It was
observed that the forecast models may be good enough to forecast growth in traffic due to
growth in socio-economic data, but are deficient in estimating the overall growth in trip
making (traffic) over the next twenty years. For this purposes time series analysis of growth
in vehicle fleet, traffic data and related changes in GDP and other economic indicators were
carried out. This process produced overall traffic growth levels by province. The future year
trip ends estimated from the regression models, as shown above, were then controlled to the
overall traffic growth volumes. Similarly, the trip ends for the two external zones (Bali and
Sumatra) were computed exogenously. This process is detailed in Appendix H.
For motorcycles and Angkots, as there were no trip matrices for these vehicle types, the
province-based growth factors were directly applied at each link level to estimate future
traffic volumes. The overall traffic growth factors for all vehicle categories and by province
are tabulated below for the years 2010 and 2020.
Table 6.1: Traffic Growth Factors 2000-2020 by Province and Vehicle Type
Car
Province /
Area

Goods Vehs.

Buses

Angkots

Motorcycles

20002010

20002020

20002010

20002020

20002010

20002020

20002010

20002020

20002010

20002020

Jakarta &
West Java

2.18

4.59

1.69

3.13

1.62

2.92

1.11

1.15

2.18

4.59

Cent. Java
& Yogya

1.86

2.95

1.51

2.10

1.46

1.98

1.05

1.08

1.86

2.95

East Java

1.72

3.09

1.38

2.18

1.33

2.04

1.05

1.10

1.72

3.09

Java Total

1.93

3.91

1.52

2.70

1.46

2.52

1.07

1.12

1.93

3.91

Bali

2000 2010 All Vehs. = 1.71

2000 2020 All Vehs. = 3.09

Sumatra

2000 2010 All Vehs. = 1.72

2000 2020 All Vehs. = 3.05

Carl Bro International a|s and Associates

34

Java Arterial Road Network Study (JARNS)

6.3

Technical ReportNo.10:
Network Model Development and Application

Future Travel Demand Matrices

The estimation of the future trip matrices was based on a doubly constraint Furness
procedure. This process takes the base year trip matrix and future year trip ends, and
iteratively adjusts the cell values until the output matrix row and column totals are equal to
the input future year trip ends.
The base year and the output future trip totals for each vehicle category are summarised
below in Table 6.2. The average peak hour matrices were then computed using the factors
derived from the base year data. The peak traffic volume throughout the day is almost
uniform, indicating that further peak spreading is unlikely. Therefore, the base year peak
period factors were retained as unchanged, along with the PCE factors. The average peak
hour trip totals in PCUs are also given in Table 6.2.
Table 6.2: Trip Totals for the Years 2000, 2010 and 2020 by Vehicle Categories
Vehicle Category

2000
24-Hour Trips
Totals (AADT)

2010
24-Hour Trips
Totals (AADT)

2020
24-Hour Trips
Totals (AADT)

Car

350,400

661,400

1,342,500

LGV

114,400

168,600

300,100

MGV

157,400

233,700

416,000

HGV

38,200

55,500

99,400

Buses

45,500

64,700

111,900

705,900

1,183,900

2,269,900

1.00

1.67

3.21

48,300

81,000

153,300

1.00

1.61

3.05

Total Vehicles
Growth Over 2000
Average Peak Hour
Total PCUs
Growth Over 2000
Source: Consultants Estimates

6.4

Forecast Future Pre-Load Traffic Volumes

Motorcycles and Angkot traffic was assigned to the network as pre-load volumes on the
strategic links. The forecast year (2010 and 2020) traffic growth rate estimates for these two
vehicle types are given in Table 6.1 above. The growth factors were directly applied to the
base year volumes on each link. No diversion of these vehicles to the new or upgraded routes
was computed. It was implied that these vehicles make only short distance journey along
specific routes, and it would be unlikely that they would divert to new routes due to network
upgrades.

Carl Bro International a|s and Associates

35

Java Arterial Road Network Study (JARNS)

7.

Technical ReportNo.10:
Network Model Development and Application

NETWORK MODEL APPLICATION

7.1

Introduction

This Section briefly describes the model application for the forecast years strategy testing.
The model outputs are defined, and the use of these outputs in the overall evaluation
framework is outlined. The model was used to test different network configurations,
incorporating numerous network development policies, strategies and schemes. The network
scenario testing is summarised in Section 6 of the main report.

7.2

Development of Future Year Networks

7.2.1

Base 2005 Network

The first step towards developing the future year networks was to take account of all the
ongoing and committed network upgrade projects to the year 2005 (i.e. those under
construction, committed and planned to be executed within the next five years). It was
necessary to include these projects before future year testing is carried out. This information
was collated from numerous project reports and the timing of the implementation of the
recommended network upgrades was confirmed with the relevant agencies. The main sources
of information were:

North Java Transport Corridor Study (NJTCS),

SURIP 1A,
Heavily Loaded Road Improvement Project HLRIP, and
Capacity Expansion Programme CAPEX 2

Around 522 km of network upgrades of various types were identified for inclusion in the
2005 network. These network upgrades are summarised below, and shown in Figure 7.1.
Further details are given in Table 7.1.

36 km of upgrade (2U7) to 7m wide single carriageway;

410 km of upgrade (4D) to 4-lane divided carriageways,
42 km of new roads, mainly 4-lane divided bypasses, and
34 km of JAGORAWI Toll road southern section widening to 6-lanes.

Table 7.1: 2000 to 2005 Network Upgrade Summary Table

2-Lane 7m

Corridor
Jakarta - Merak
North Java Corridor Completed
North Java Corridor Committed
North Java Corridor Bypass
Bekasi Cibinong
Cikempek Pedalarang
Bandung Cirebon
Yogyakarta Surabaya Completed
Yogyakarta Surabaya Committed

Carl Bro International a|s and Associates

Online
(km)
35.6
-

4-Lane
Divided

4-Lane
Divided

Onlin Bypass
(km)
(km)
8.5
6.5
217.9
39.1
36.0
62.0
16.8
62.6
-

Total
(km)
8.5
6.5
217.9
39.1
36.0
35.6
62.0
16.8
62.6

Total
Cons.
Costs(1)
48.5
(2)
1,215.8
623.1
204.5
32.4
352.1
(2)
349.1

36

Merak(

JAKARTA

Serang(

>

)
)

Cikampek
)

)
)

)
)

Bogor.

)
)

Sukabumi.

BANDUNG.

Tegal
)

Rembang.

Cirebon.

Pekalongan
(

SEMARANG.

Tuban

)
)

Sumenep.

)
)
)

Salatiga.

Ciamis.

Magelang.

Cilacap.

Jombang

Surakarta

n
SURABAYA

YOGYAKARTA

)
)

)
)

( Gempol
)

Madiun

Probolinggo

)
)

)
)

(Malang.

Tulungagung

)
)

)
)

Banyuwangi.

Pacitan

LEGEND :
Upgrade to 2U7
Upgrade to 4D
New By-pass
Toll Road Upgrade to Dual-3

FIGURE 7.1: NETWORK UPGRADE 2000 TO 2005

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

Table 7.1: 2000 to 2005 Network Upgrade Summary Table

2-Lane 7m

Online
(km)

Corridor
Yogyakarta Surabaya Bypass
Sub-total Non-Toll Road Upgrades

4-Lane
Divided

Onlin
(km)
-

4-Lane
Divided

Bypass
(km)
3.1

Total
(km)

Total
Cons.
Costs(1)

3.1

48.2

35.6

410.3

42.2

488.1

2,873.7

Jagorawi Upgrade to Dual-3

33.5

239.5

Total 2000 to 2005 Upgrades

35.6

410.3

42.2

521.6

3,113.2

7.2.2

Future Year 2010 Networks

The 2005 network formed the basis for testing future year network scenarios. The network
upgrade options were:

Upgrade an existing road to 7m wide single carriageway (2U7);

Upgrade an existing road to 14m wide 4-lane divided carriageway (4D);
Upgrade of an existing toll road of 4-lanes to 6-lanes; and
The construction of new toll roads along inter-urban corridors.

Once an upgrade of the network corridor/ section(s) (irrespective of how small or large) need
to be evaluated, the network database update procedures were applied to prepare a new
network computer file for input to the TRIPS network assignment model. The following key
items were computed and updated in the new test network computer file:

Link Capacity;
Free Flow Speed;
Speed Flow Curve allocation based on the new free flow speed;
Revised road classification for the calculation of accidents.

The 2010 network upgrades used the 2005 network as a base case network. All upgrades
were applied only to the strategic inter-urban links and the toll roads. The capacity of the
remainder of the network (urban/ Kota links, notional urban corridors and toll road access
links) was kept at the same level as it was in the 2005 base network. It was implied that the
capacity expansion in the urban area networks by traffic management and other local
measures would continue to provide adequate additional capacity for the inter-urban traffic
on these urban area links and corridors.
7.2.3

Future Year 2020 Networks

The 2020 network upgrades used the 2010 Do-Something network as the base network. Once
again, all future test scenarios were applied only to the strategic inter-urban links and the toll
roads. The capacity of the remainder of the network (urban/ Kota links, notional urban
corridors and toll road access links) was investigated by assigning the 2020 demand to the
2010 Do-Something network. It was found that these urban area links and corridor would
not have adequate capacity to effectively accommodate the 2020 inter-urban traffic demand.
As a result, it was estimated that an increase of 100 percent in the capacity of urban links
would be required to maintain a reasonable level of congestion-free speeds for the inter-urban
traffic through the urban areas. Similarly, an increase of 50 percent in the capacity of the
inter-urban toll road access links would be required to accommodate the 2020 toll road traffic
volumes.

Carl Bro International a|s and Associates

37

Java Arterial Road Network Study (JARNS)

7.3

Technical ReportNo.10:
Network Model Development and Application

Future Year Network Model Parameters

The other key data items that need to be defined for the use of the network model for the
2010 and 2020 forecast years were the GC parameters. These were forecast from the base
year data by increasing the value of time by 4 percent per annum in real terms. Similarly, the
future values of vehicle operating costs and the perceived level of tolls and value of time for
toll road users were estimated. These parameters are summarised in Table 7.2 below.
Table 7.2: Network Model Traffic Assignment Parameters for 2000, 2010 and 2020
Assignment Parameter

2000

2010

2020

Value of Time (Rp / hour / PCU)

27,000

32,000

44,000

VOT for Toll Road Users (Rp / hour / PCU)

24,300

28,800

39,600

700

150

150

Tolls (Rp/km/PCU)
100
Note: All values are in year 2000 prices. Source: Consultants Estimate

150

200

Vehicle Operating Cost (Rp/km/PCU)

7.4

Testing of Future Year Network Scenarios and Evaluation

The future year scenario testing involved running the equilibrium assignment model with
the appropriate year demand and the assignment parameters. The model converged after 25
iterations for the year 2010 tests. However, due to increased congestion and assignment for
some O-D pairs oscillating between two marginally different cost routes, the 2020 tests
required 15 further iterations for the model to achieve convergence. The key model outputs
were:
1.
2.
3.
4.
5.
6.
7.
8.
9.

Total average peak hour assigned link volumes

Pre-load link volumes of motorcycles and Angkots
Sum of the above two volumes giving total network average peak hour volumes
Volume Capacity Ratio, where volume is defined above
Link volumes as defined in 1-3 above for 24-hours
Network-wide travel statistics of vehicle-hours and vehicle-kms travelled for
both average peak hour and the AADT by type of road
Network-wide travel speeds for average peak hour by type of road
Network-wide AADT-kms travelled by road classification used by the accident
evaluation module
The converged network-wide link travel times after the assignment.

For every test scenario, these network statistics were computed to investigate network
operational performance. The VCR were plotted to examine the congested elements of the
network. The changes in network-wide speed and vehicle-kms travelled were examined to
identify the impacts of network upgrades.

7.5

Network Model Inputs to the Economic Evaluation Framework

The economic evaluation process used the converged output networks from the last 4
iterations in case of 2010 tests, and the last 6 iterations in the case of 2020 tests, (both for the
base and test case scenarios) to skim the following values for each O-D pair (i.e. skim
matrices):

Travel Times;
Travel Distances;
Vehicle Operating Costs based on the converged speeds.

Carl Bro International a|s and Associates

38

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

The route choice parameters used for building the paths for skimming the above values were
the same as used in the assignment process, and are given in Table 7.1 above. The evaluation
model used the simple linear average of these skimmed matrices, to estimate the total
network wide benefits. The complete evaluation process is defined in Technical Report 11
and its use is explained in Section 6.

Carl Bro International a|s and Associates

39

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

APPENDIX A:
ZONE SYSTEM DEFINITION

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

A1. INTRODUCTION
The JARNS study divided the Island of Java and Madura into 262 traffic zones and islands of
Bali and Sumatra are represented by two external zones. Table A.1 below presents a summary
of number of zones in each region/province. Generally, every kabupaten was divided into two or
more zones, and Kotamadyas with discrete administrative boundaries were kept as single zones.
Table A.1: Number of Zones in the JARNS Study Area
Region / Province
Number and Name
1
2
3
4
5

DKI Jakarta
West Java
Central Java
Yogyakarta
East Java

Java Total

6
7

Number of
Kabupaten /
Kotamadya
5
26
35
5
37

External Zone
External Zone

Total

Number of
Zones

Sequential
Zone Numbers

5
72
93
6
86

1-5
6-77
78-170
171-176
177-262

108

262

1-262

Bali
Sumatra

1
1

263
264

264

264

A2. ZONE NUMBERING SYSTEM

The study adopted a hierarchical zone numbering system, in addition to the sequential zone
numbers. The hieratical zone numbering system was defined in such a way, so that the location of
each zone can be easily identified in relation to its parent province and Kabupaten or Kotamadya.
Therefore, each zone was given a 4-digit numeric code, say XYYZ:
Where:

X indicates the province/region number as shown in the Table A.1 above;

YY is the 2-digit Kabupaten / Kotamadya code as used by BPS;
Z is a single digit number 0-9 of the zone within a Kabupaten or Kotamadya

A full description of the zone numbers (sequential and hierarchical) and the description of the
area covered by each zone in terms of the Kecamatan within each zone are given in Table A.2.
Figures A.1a, A.1b and A.1c show the zone boundaries with the hierarchical zone numbers for the
regions of Jakarta and West Java; Central Java and Yogyakarta; and East Java respectively.

Carl Bro International a|s and Associates

A- 1

2206

2202
2201

2205

2203
2193

2204

2750

1750
1740
1730

1720
2760
1710
2191 2192

2011
2021

2710
2023

2141

2152

2038

2036

2012

2172

2171

2035
20342031

2013

2182
2181

2022

2161

2033

2063

2112
2153

2132

2122

2064

2111
2740
2121

2052
2730

2720

2142

2151

2162

2037 2039 2053

2040

2143

2154

2032

2102

2133
2131

2123

2062

2101

2051

2065

2113

2103

2072

2061
2083

2093
2091

2071

301
2082
2092
2081

LEGEND :
Zone Boundary
Province Boundary
XXXX

Hierarchical Zone Number

FIGURE A.1a: STUDY AREA ZONE PLAN

( JAKARTA & WEST JAVA )

2142

3202
3182

3201
112
3203
3192

3181

3171
3172

3183
3760
3291

3292

3271

3282
3281

3272

3211

3750
3252

3283

3213

3253

3273

3161

3221

3042

3032
3021
3013

3023

3012
3011

3031
3033
3022

3232

3041

3152

3043

3053

3082

3072

3730
3223

3092

3061

3091

3062

4040

3142
3141

3133
3720
3131
3112

3085

3051
3063

3163

3143

3083

3710
3084
3081

3052

3153

3222
3231

3071

3073

3162

3151

3154

3
3293

3184

3164

3740

3242

3260

3191

3214

3241

3251

3212

3173

3101
3103

3132

3111

3102

3113
4010

4710
3121

4020
4031

3122

4032
3123

LEGEND :
Zone Boundary
Province Boundary
XXXX

Hierarchical Zone Number

FIGURE A.1b: STUDY AREA ZONE PLAN

( CENTRAL JAVA & YOGYAKARTA )

5233

5242

5231

5252

5232

5261

5290
5262

5270

5280

5251
5241

5243
5223

5213

5221

5222

5191

5182

5760

5154

5162

5171

5181

5780

5152

5174

5211 5212

5202 5770
5192
5203
5201

5253

5173

5172

5151
5153

5161
5141

5121

5061
5710

5062

5021

5142
5075

5074

5063
5022

5013

5011

5740

5133

5032

5123

5132

5112
5113

5051

5082

5071

5720

5081

5052
5072

5042
5073

5122

5111

5730

5031
5041

5014

5143

5131

5043
5012

5750

5101

5091
5092

5083
5093

5102

LEGEND :
Zone Boundary
Province Boundary
XXXX

Hierarchical Zone Number

FIGURE A.1c: STUDY AREA ZONE PLAN

(EAST JAVA)

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

1
2
3
4
5
6

Hier.
No.

1710
1720
1730
1740
1750
2011

Zone Name

Jakarta South
Jakarta East
Jakarta Central
Jakarta West
Jakarta North
Pandeglang

7 2012

Rangkasbitung

Jakarta South
Jakarta East
Jakarta Central
Jakarta West

South Jakarta city.

East Jakarta city.
Central Jakarta city.
West Jakarta city.

Jakarta North

North Jakarta city.

Pandeglang

Lebak

Bogor

Pandeglang, Cadasari, Banjar.

Saketi, Bojong, Munjul, Cikeusik,
Mandalawangi, Cimanuk.
Labuan, Jiput, Menes, Pagelaran,
Cigeulis, Sumur, Cimanggu, Cibaliung,
Panimbang, Picung.
Rangkasbitung,
Warunggunung,
Cimarga, Cibadak.
Cipanas, Sajira, Muncang, Bayah,
Panggarangan, Maja, Cibeber, Cijaku,
Cikulur.
Malingping, Gunungkencana, Bojong
Manik, Cileles, Leuwidamar, Banjarsari.
Cibinong.
Cileungsi, Cariu, Jonggol, Gunung Putri.
Citeureup,
Babakan
Medang,
Sukamakmur.
Bojonggede, Semplak.
Parung, Gunung Sindur, Rumpin, Parung
Panjang.
Leuwiliang,
Ciampea,
Jasinga,
Nanggung, Darmaga, Tenjo, Cigudeg,
Cibunggulang, Ciomas, Kedunghalang.
Ciawi, Cijeruk, Caringin, Pamijahan.
Kota Depok, Cimanggis, Sawangan,
Limo, Pancoran Mas, Beji, Sukmajaya
Cisarua, Megamendung.
Cibadak,
Nagrak,
Parungkuda,
Cikembar, Warung Kiara, Pel. Ratu,
Lengkong, Jampang Tengah, Jampang
Kulon, Surade, Cisolok, Cikidang,
Kalapanunggal, Tegalbuleud, Ciomas,
Kalibunder, Parakan Salak, Ciracap,
Cidolog, Kabandungan, Kadudampit,
Pabuaran, Cidahu.
Cianjur, Warungkondang, Cibeber,
Agrabinta,
Cempaka,
Sukanegara,
Pagelaran, Kadupandak, Naringgul,
Cibinong, Sindangbarang, Cidaun,
Cikalong Kulon, Tanggeung, Takokak.

Labuan

9 2021

Scope of Area
- Kecamatan

1
2
3
4
5
1

Saketi

8 2013

Kabupaten /
Kotamadya
Code
Name

10 2022

Cipanas

11 2023

Malingping

12 2031
13 2032

Cibinong
Cileungsi

14 2033

Citeureup

15 2034

Bojonggede

16 2035

Parung

17 2036

Leuwiliang

18 2037

Ciawi

19 2038

Depok

20 2039

Cisarua

21 2040

Cibadak

Sukabumi

22 2051

Cianjur

Cianjur

Carl Bro International a|s and Associates

A- 2

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

23 2052

Ciranjang

24 2053

Cipanas

25 2061

Soreang

26 2062

Kabupaten /
Kotamadya
Code
Name

Cimahi

Bandung

Garut

Tasikmalaya

Ciamis

27 2063 Padalarang
28 2064

Lembang

29 2065

Majalaya

30 2071

Garut

31 2072

Balubur
Limbangan

32 2081

Tasikmalaya

33 2082

Singaparna

34 2083

Cisayong

35 2091

Ciamis

36 2092

Banjar

Carl Bro International a|s and Associates

Scope of Area
- Kecamatan
Ciranjang,
Bojongpicung,
Mande,
Karang, Tengah, Sukaresmi, Cilaku,
Sukaluyu.
Pacet, Cugenang.
Soreang, Banjaran, Pasir Jambu,
Arjasari,
Ciwidey,
Pengalengan,
Pameungpeuk, Katapang, Dayeuhkolot,
Margaasih, Baleendah, Bojongsoang,
Margahayu, Cilengkrang.
Cimahi Selatan, Cimahi Tengah, Cimahi
Utara, Batujajar, Cililin, Sindang Kerta,
Gunung Halu.
Padalarang,
Cipatat,
Cipeundeuy,
Cikalong Wetan, Ngamprah.
Lembang,
Cisarua,
Cipongkor,
Parompong.
Majalaya, Paseh, Pacet, Ibun, Kertasari,
Ciparay,
Cicalengka,
Cileunyi,
Cimenyan,
Cikancung,
Cimaung,
Rancaekek.
Garut, Cilawu, Bayongbong, Cisurupan,
Pendeuy,
Cikajang,
Pakenjeng,
Bungbulang,
Cisewu,
Pamulihan,
Cisompet,
Pameungpeuk,
Cikelet,
Singajaya, Cibiuk, Karangpawitan,
Wanaraja, Banyuresmi, Sukawening,
Tarogong, Samarang, Selawi, Talegong,
Cibalong.
Balubur
Limbangan,
Malangbong,
Cibatu, Kadungora, Leles, Leuwigoong,
Banjar Wangi.
Kotif Tasikmalaya, Sukaraja, Cibalong,
Karang Nunggal, Bantar Kalong, Salopa,
Cikatomas,
Cikalong,
Kawali,
Cibeureum, Manon Jaya, Cineam,
Cipatujah, Pangatengah, Tanjung Jaya,
Jamanis, Cipedes, Cihideung, Tawang.
Singaparna, Leuwisari, Cigalontang,
Salawu, Taraju, Sodong Hilir, Bojong
Gambir.
Cisayong, Indihiang, Rajapolah, Ciawi,
Pagerageung.
Ciamis,
Cikoneng,
Cijeungjing,
Sukadana, Jatinegara.
Banjar, Cimaragas, Cisaga, Pamarican,
Banjarsari,
Lakbok,
Padaherang,
Kalipucang,
Pangandaran,
Parigi,
Cijulang, Langkaplancar, Cigugur,
Pataruman, Langensari, Purworaharja,

A- 3

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

37 2093

Kawali

38 2101

Kuningan

39 2102

Cilimus

40 2103

Ciawigebang

41 2111

Sumber

42 2112

Palimanan

43 2113

Losari

44 2121

Majalengka

45 2122

Kadipaten

46 2123

Cikijing

47 2131

Sumedang

48 2132

Cimalaka

49 2133

Tanjung Sari

50 2141
51 2142

Indramayu
Karang Ampel

52 2143

Lohbener

53 2151

Subang

54 2152

Pamanukan

55 2153

Sagalaherang

56 2154

Kabupaten /
Kotamadya
Code
Name

Kalijati

10

Kuningan

11

Cirebon

12

Majalengka

13

Sumedang

14

Indramayu

15

Subang

Carl Bro International a|s and Associates

Scope of Area
- Kecamatan
Cimerak, Tambak Sari, Cidolok,
Sidomulih.
Kawali, Cipaku, Panjalu, Panawangan,
Rajadesa,
Rancah,
Sadananya,
Cihaurbeuti, Panumbangan.
Kuningan, Kadugede, Ciniru, Cibubur,
Darma, Selajambe, Kramatmulya.
Cilimus,
Jalaksana,
Mandiracan,
Pasawahan.
Ciawigebang, Cidahu, Lebakwangi,
Luragung, Garwangi, Cibingbin, Ciwaru,
Subang.
Sumber,
Beber,
Sedong,
Weru,
Kapetakan.
Palimanan,
Klangenan,
Plumbon,
Susukan,
Arjawinangun,
Gegesik,
Ciwaringin.
Losari, Babakan, Ciledug, Waled,
Lemahabang,
Karang
Sambung,
Astanajapura, Mundu.
Majalengka, Sukahaji, Rajagaluh, Leuwi
Munding, Sindang Wangi, Cigasong,
Panyingkiran.
Kadipaten, Dawuan, Jatiwangi, Sumber
Jaya, Kertajati, Jatitujuh, Ligung,
Palasah.
Talaga, Maja, Baturajeg, Lemah, Sugih,
Cikijing, Argapura.
Sumedang, Situraja, Darmaraja, Wado,
Cadas Ngampar, Cibugel.
Cimalaka, Conggeang, Tomo, Buah Dua,
Paseh, Tanjung Kerta, Ujung Jaya.
Tanjung Sari, Cikeruh, Rancakalong,
Cimanggung.
Indramayu, Balongan.
Karang Ampel, Jatinyuat, Krangkeng.
Lohbener,
Jatibarang,
Cikedung,
Anjatan,
Haurgeulis,
Gabuswetan,
Kandanghaur, Losarang, Sindang, Lelea,
Bangodua, Kertasemaya, Sliyeg, Bongas,
Widasari, Sukra, Kroya.
Subang, Cibogo, Cipunagara.
Pamanukan, Ciasem, Pusakanagara,
Binong, Pagaden, Tanjung Siang,
Sompreng, Blanakan, Cijambe, Cikaum.
Sagalaherang, Cisalak, Jalan Cagak.
Kalijati,
Purwadadi,
Pabuaran,
Cipeundeuy, Patok Beusi, Legon Kulon.

A- 4

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

57 2161

Purwakarta

58 2162

Karawang

60 2172

Cibitung

62 2182

Serpong
Ciputat

65 2193

Serang

67 2202

Bogor
Sukabumi
Bandung
Cirebon
Tangerang
Bekasi
Cilacap

20

Serang

71
72
73
74
75
76
1

Bogor
Sukabumi
Bandung
Cirebon
Tangerang
Bekasi

Kota Bekasi

Cilacap

Kotif Cilacap.
Kroya, Binangun, Nusawungu, Adipala,
Maos, Kesugihan, Sampang.
Majenang,
Dayehluhur,
Wanareja,
Cimanggu, Karang Pucung, Sidareja,
Kadungreja, Gandrungmangu, Cipari,
Patimuan, Kawungganten, Jeruklegi.

Cilegon

72
73
74
75
76
77
78

Tangerang

Anyer

71 2206

19

Baros

70 2205

Bekasi

Cikande

69 2204

18

Kramatwatu

68 2203

Karawang

Cikupa

66 2201

17

Purwakarta,
Jatiluhur,
Pasawahan,
Campaka
Plered, Darangdan, Wanayasa, Bojong,
Tegalwangi, Sukatani.
Karawang, Rengasdengklok, Teluk
Jambe, Pangkalan, Pedes, Cibuaya, Batu
Jaya, Rawamerta, Telagasari, Pakis Jaya.
Cikampek,
Jatisari,
Lemahabang,
Cilamaya,
Klari,
Kertawaluya,
Tempuran,
Tirtamulya,
Tirtajaya,
Ciampel.
Cibitung, Setu, Tambun, Pebayuran,
Tambelang.
Cikarang, Cibarusah, Lemahabang,
Sukatani, Cabang Bungin, Serang,
Muara Gembong, Kedung Waringin.
Serpong, Cisauk
Ciputat, Pondok Aren, Pamulang
Cikupa, Tigaraksa, Curug, Legog,
Pasarkemis, Balaraja, Kresek, Kronjo,
Mauk, Rajeg, Paku Haji, Pagedangan,
Panongan
Serang
Kramatwatu,
Waringin
Kurung,
Taktakan, Kasemen, Bojonegoro
Cikande, Ciruas, Walantaka, Kragilan,
Pontang,
Tirtayasa,
Carenang,
Pamarayan, Kopo, Cikeusa, Jawilan.
Baros, Ciomas, Petir, Pabuaran, Curug,
Cipocah, Cipocok Jaya.
Anyer, Cinangka, Mancah, Padaruncang
Kota Cilegon, Cibeber, Pulomerak,
Cimanda
Kota Bogor
Kota Sukabumi
Kota Bandung,
Kota Cirebon
Kota Tangerang,

Cikarang

63 2191
64 2192

Purwakarta

Cikampek

61 2181

2710
2720
2730
2740
2750
2760
3011

79 3012

Kroya

80 3013

Scope of Area
- Kecamatan

16

Kab. Purwakarta

59 2171

Kabupaten /
Kotamadya
Code
Name

Majenang

Carl Bro International a|s and Associates

A- 5

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

81 3021

Purwokerto

82 3022

Purbalingga

85 3032

Banjarnegara

88 3042

Kebumen

91 3052

Purworejo

94 3062

Wonosobo

97 3072

Mungkid
Secang
Pakis

102 3084

Magelang

Borobudur

103 3085

Wonosobo

Leksono

99 3081
100 3082
101 3083

Kretek

98 3073

Purworejo

Kutoarjo

96 3071

Bagelen

95 3063

Kebumen

Gombong

93 3061

Prembun

92 3053

Banjarnegara

Mandiraja

90 3051

Wanayasa

89 3043

Purbalingga

Kemangkon

87 3041

Bobotsari

86 3033

Banyumas

Wangon

84 3031

Banyumas

83 3023

Kabupaten /
Kotamadya
Code
Name

Muntilan

Carl Bro International a|s and Associates

Scope of Area
- Kecamatan
Kotif Purwokerto, Patikraja, Baturaden,
Karanglewas, Kedung Banteng.
Banyumas,
Sokaraja,
Kalibagor,
Kembaran,
Sumbang,
Somagede,
Kebasen, Tambak, Kemranjen, Sumpiuh.
Ajibarang,
Cilongok,
Wangon,
Purwojati, Jatilawang, Rawalo, Lumbir,
Gumelar, Pekuncen.
Purbalingga, Kalimanah, Kaligondang,
Padamara.
Bobotsari,
Karanganyar,
Mrebet,
Kutasari, Karangmoncol, Rembang,
Bojongsari.
Kemangkon,
Bukateja,
Kejobong,
Pengadegan.
Banjarnegara, Banjarmangu, Wanadadi,
Punggelan, Rakit, Madukora, Sigaluh.
Wanayasa, Kalibening, Karangkobar,
Pejawaran, Pagetan, Batur.
Mandiraja, Purwonegoro, Bawang,
Susukan, Klampok.
Kebumen, Pejagoan, Sadang, Klirong,
Alian.
Prembun, Kutowinangun, Mirit, Bulu
Pesantren, Ambal.
Gombong,
Sempor,
Kuwarasan,
Karanggayam, Karanganyar, Rowokele,
Adimulyo, Sruweng, Puring, Ayah,
Petanahan, Buayan.
Purworejo, Gebang, Bayan, Loano,
Bener, Banyu Urip.
Bagelen,
Purwodadi,
Ngombol,
Kaligesing.
Kutoarjo, Butuh, Kemiri, Bruno, Grabag,
Pituruh.
Wonosobo, Mojo Tengah, Garung,
Kejajar.
Kretek, Kalijajar, Sapuran, Kepil.
Leksono,
Selomerto,
Kaliwira,
Wadaslintang, Watumalang.
Mungkid, Sawangan.
Secang, Grabag.
Pakis, Tegalrejo, Ngablak, Candimulyo.
Borobudur,
Salaman,
Kaliangkrik,
Kajoran, Tempuran.
Muntilan, Dukun, Srumbung, Salam,
Ngluwar.

A- 6

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

104 3091

Boyolali

105 3092

Klaten

107 3102

Sukoharjo

110 3112
111 3113

Wonogiri

113 3122

Karanganyar

116 3132

Sragen

119 3142

Purwodadi

122 3152

Sragen

15

Grobogan

Toroh

124 3154

14

Wirosari

123 3153

Karanganyar

Tanon

121 3151

13

Sambungmacan

120 3143

Wonogiri

Jaten

118 3141

12

Karangpandan

117 3133

Sukoharjo

Giriwoyo

115 3131

11

Jatisrono

114 3123

Klaten

Grogol
Weru

112 3121

10

Manisrenggo

109 3111

Boyolali

Delanggu

108 3103

Banyudono

106 3101

Kabupaten /
Kotamadya
Code
Name

Godong

Carl Bro International a|s and Associates

Scope of Area
- Kecamatan
Boyolali,
Musuk,
Cepogo,
Selo,
Mojosongo, Teras, Ampel.
Banyudono, Ngemplak, Sambi, Simao,
Karanggede, Wonosegoro, Kemusu,
Klego, Juwangi, Sawit, Nogosari,
Andong.
Klaten, Jatinom, Tulung, Ngawen.
Delanggu,
Polanharjo,
Wonosari,
Juwiring,
Ceper,
Karanganom,
Karangdowo, Pedan, Bayat, Trucuk,
Kalikoter, Cawas.
Manisrenggo, Prambanan, Gantiwarno,
Wedi,
Jogonalan,
Kebunarum,
Kemalang, Karangnongko.
Sukoharjo,
Bendosari,
Nguter,
Mojolaban, Polokarto.
Grogol, Baki, Kartosuro, Gatak
Tawangsari, Weru, Bulu.
Wonogiri, Manyaran, Wuryantoro,
Selogiri.
Jatisrono,
Sidoharjo,
Ngadirejo,
Girimanto,
Jatipurno,
Bulukerto,
Slogohimo, Purwantoro, Kismantoro,
Jatiroto, Nguntoronadi.
Giriwoyo, Giritontro, Pracimantoro,
Baturetno, Batuwarno, Paranggupito,
Tirtomoyo, Eromoko, Karang Tengah.
Karanganyar.
Karangpandan, Jenawi, Ngargoyoso,
Tawangmangu, Jumantoro, Matesih,
Jumapolo,
Jatiyoso,
Jatipuro,
Mojogendang, Kerjo.
Jaten,
Tasikmadu,
Gondangrejo,
Kebakramat, Colomadu.
Sragen, Karangmalang, Kedawung.
Sambungmacan, Gondang, Ngrampal,
Tangen, Gesi, Jenar, Sambirejo,
Sidoharjo, Sukodono.
Masaran,
Tanon,
Sumberlawang,
Mondokan,
Gemolong,
Plupuh,
Kalijambe, Miri.
Purwodadi, Grobogan, Brati.
Wirosari, Tawangharjo, Ngarungan,
Kradenan, Gabus, Pulokulon.
Toroh, Geyer.
Godong, Penawangan, Karangrayung,
Gubug, Kedungjati, Klambu, Tegowanu,

A- 7

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

Kabupaten /
Kotamadya
Code
Name

Scope of Area
- Kecamatan
Tanggungharjo.

125 3161
126 3162

Blora
Cepu

127 3163

Randublatung

128 3164

Kunduran

129 3171

Rembang

130 3172

Lasem

131 3173
132 3181

Sulang
Pati

133 3182

Tayu

134 3183

16

Juana

135
136
137
138
139

3184
3191
3192
3201
3202

Kayen
Kudus
Jekulo
Jepara
Bangsri

Blora

17

Rembang

18

Pati

19

Kudus

20

Jepara

140 3203

Pacangaan

141 3211

Demak

21

Demak

Gajah
Mranggen
Karang Tengah
Ungaran

22

Semarang.

23

Temanggung

24

Kendal

142
143
144
145

3212
3213
3214
3221

146 3222

Bawen

147 3223

Tengaran

148 3231

Temanggung

149 3232

Parakan

150 3241

Kendal

151 3242

Sukorejo

Carl Bro International a|s and Associates

Blora, Tunjungan.
Cepu, Jepon, Jiken, Sambong.
Randublatung,
Jati,
Kedungtuban,
Kradenan.
Kunduran, Ngawen, Banjar Rejo,
Todanan, Bogorejo, Japah.
Rembang, Kaliori.
Lasem, Sluke, Pancur, Pamotan, Sedan,
Kragan, Sarang, Sale.
Sulang, Bulu, Genem, Sumber.
Pati, Wedarijekso, Gembong, Margarejo.
Tayu,
Dukuh
Sekti,
Cluwak,
Gunungwungkal,
Margoyoso,
Tlogowungu, Trangkil.
Juana, Jakenan, Batangan, Jaken,
Winong, Puncakwangi.
Kayen, Tambakromo, Sukolilo, Gabus.
Kudus, Jati, Undaan, Kaliwungu.
Jekulo, Mejobo, Gebog, Dawe, Bae.
Jepara, Tahunan.
Bangsri, Keling, Mlongo, Karimunjawa
Pacangaan, Mayong, Kedung, Welahan,
Batealit, Nalumsari.
Demak, Bonang, Wedung, Dempet,
Wonosalam.
Gajah, Karanganyar, Mijen.
Mranggen, Karangawen.
Karang Tengah, Sayung, Guntur.
Ungaran.
Bawen, Ambarawa, Jambu, Beringin,
Tuntang,
Banyubiru,
Sumowono,
Pabelan, Bergas, Pringapus.
Tengaran, Suruh, Susukan, Getasan.
Temanggung,
Tembarak,
Kaloran,
Kandangan, Kranggan.
Parakan, Bulu, Kedu, Ngadirejo, Jumo,
Candiroto, Tretep, Pringsurat.
Kendal, Pegandom, Patebon, Brangsong,
Kaliwungu, Cepiring.
Sukorejo, Patean, Singorojo, Boja,
Plantungan,
Pagerruyung,
Weleri,
Gemuh,
Limbangan,
Rowosari,
Kangkung.

A- 8

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

Kabupaten /
Kotamadya
Code
Name
Batang
25

Scope of Area
- Kecamatan

152 3251
153 3252

Batang
Subah

154 3253

Bandar

155 3260

Kedungwuni

26

Pekalongan

156 3271

Pemalang

27

Pemalang

157 3272

Petarukan

158 3273

Randudongkal

159 3281

Slawi

28

Tegal

160 3282

Suradadi

161 3283

Margasari

162 3291

Brebes

29

Brebes

163 3292

Ketanggungan

164 3293

Bumiayu

165
166
167
168
169
170

Magelang
Surakarta
Salatiga
Semarang
Pekalongan
Tegal

71
72
73
74
75
76

Magelang
Surakarta
Salatiga
Semarang
Pekalongan

Batang, Warungasem, Tulis.

Subah, Gringsing, Limpung, Tersono.
Bandar, Wonotunggal, Blado, Bawang,
Reban.
Kedungwuni, Wonopringgo, Bojong,
Kajen, Kesesi, Karanganyar, Doro,
Talun,
Sragi,
Paninggaran,
Petungkriyono,
Lebakbarang,
Kandangserang.
Pemalang, Taman
Petarukan,
Ulujami,
Comal,
Ampelgading, Bodeh.
Randudongkal, Bantarbolang, Moga,
Watukumpul, Pulosari, Belik.
Slawi, Pangkah, Kedung Banteng,
Adiwerna, Dukuh Waru.
Suradadi, Wanurejo, Kramat.
Margasari, Balapulang, Lebak Siu, Bumi
Jawa, Bojong, Pagerbarang, Jatinegara.
Brebes, Jatibarang, Wanasari, Songgom
Ketanggungan,
Tanjung,
Kersana,
Bulakamba, Banjarharjo, Larangan,
Losari.
Bumiayu, Tonjong, Bantar Kawung,
Salem, Paguyangan, Sirampog.
Kota Magelang.
Kota Surakarta.
Kota Salatiga.
Kota Semarang.
Kota Pekalongan.

Tegal

Kota Tegal.

171 4010

Wates

Kulonprogo

172 4020

Bantul

Bantul

173 4031

Wonosari

Gunungkidul

174 4032

Semanu

175 4040

Sleman

Sleman

3710
3720
3730
3740
3750
3760

Carl Bro International a|s and Associates

Wates, Temon, Panjatan, Galur, Lendah,

Sentolo, Pengasih, Kokap, Girimulyo,
Nanggulan, Samigaluh, Kalibawang.
Bantul, Srandakan, Sanden, Kretek,
Pundong, Bambang Lipuro, Pandak,
Panjangan, Jetis, Imogiri, Dlingo, Pleret,
Sewon.
Wonosari, Panggang, Paliyan, Tepus,
Playen, Patuk, Nglipar, Saptosari,
Gedangsari.
Semanu,
Rongkop,
Ponjong,
Karangmojo, Ngawen, Semin.
Sleman, Godean, Moyudan, Minggir,
Seyegan, Tempel.

A- 9

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

Kabupaten /
Kotamadya
Code
Name

176 4710

Yogyakarta

71

177
178
179
180

Pacitan
Tegalombo
Ngadirejo
Punung

181 5021

Ponorogo

182 5022

Balong

183 5031

Trenggalek

184 5032

Dongko

185 5041

Tulungagung

186 5042

Ngunut

187 5043

Gondang

188 5051

Srengat

189 5052

Wlingi

190 5061

Tarokan

191 5062

Pare

192 5063

Kraas

193 5071

Kepanjen

194 5072

Dampit

195 5073

Sumber Pucung

196 5074

Batu

5011
5012
5013
5014

Carl Bro International a|s and Associates

Scope of Area
- Kecamatan

Kota
Yogyakarta,
Banguntapan,
Piyungan, Kasihan, Sedayu (Kab.
Bantul), Gamping, Mlati, Depok,
Yogyakarta
Berbah, Prambanan, Kalasan, Ngemplak,
Ngaglik, Turi, Pakem, Cangkringan
(Kab. Sleman).
Pacitan
Pacitan, Kebonagung
Tegalombo, Arjosari, Nawangan, Bandar
Ngadirejo, Sudimoro, Tulakan
Punung, Donorejo, Pringkuku
Ponorogo, Siman, Pulung, Sooko,
Ponorogo
Babadan, Sampung, Sukorejo, Badegan,
Jenangan, Ngebel, Kauman, Jambon
Balong, Jetis, Sambit, Mlarak, Sawoo,
Bungkal, Slahung
Trenggalek, Bendungan, Pogalan, Tugu,
Trenggalek
Karangan, Duroman
Dongko, Panggul, Kampak, Munjungan,
Watulino, Gandusari, Pule
Tulungagung, Kedungwaru, Basuki,
Bandung,
Pakel,
Campur
darat,
Tulungangung
Bonyolangu,
Ngantru,
Tanggung
Gunung, Sumbergempol, Kauman
Ngunut,
Rejotangan,
Kalidawir,
Pucanglaban
Gondang, Karang rejo, Pagerwojo,
Sendang
Blitar
Srengat, Udanawu, Ponggok, Wonodadi.
Wlingi, Talun, Gandusari, Doko,
Kesamben, Selorejo, Binangun, Wates,
Panggungrejo, Selopuro.
Tarokan, Papar, Purwosari, Plemahan,
Kediri
Banyakan.
Pare, Kandangan, Gurah, Plosoklaten,
Puncu, Pagu, Kepung, Kunjang, Ringin
Rejo.
Kraas, Ngadiluwih, Kandat, Wates,
Ngancar
Malang
Kepanjen, Ngajum, Pagak, Wonosari.
Dampit,
Ampelgading,
Turen,
Sumbermanjing Wetan, Gondanglegi,
Bantur, Bululawang, Wajak, Tajinan,
Poncokusumo, Pagelaran, Gedangan,
Tirtoyudo.
Sumber Pucung, Kalipare, Donomulyo,
Kromengan.
Kotip
Batu,
Pujon,
Ngantang,

A- 10

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

Kabupaten /
Kotamadya
Code
Name

Scope of Area
- Kecamatan
Kasembon, Junrejo, Bumiaji.

197 5075

Singosari

198 5081

Lumajang

199 5082

Klakah

200 5083

Pasirian

201 5091

Jember

202 5092

Silo

203 5093

Bangsalsari

204 5101

Banyuwangi

205 5102

Rogojampi

206 5111

Bondowoso

207 5112

Klabang

208 5113

Maesan

209 5121

Situbondo

210 5122

Asembagus

211 5123

Besuki

212 5131

Leces

213 5132

Kraksaan

214 5133

Tongas

215 5141

Gempol

216 5142

Purwosari

Lumajang

Jember

10

Banyuwangi

11

Bondowoso

12

Situbondo

13

Probolinggo

14

Pasuruan

Carl Bro International a|s and Associates

Singosari, Lawang, Karangploso.

Lumajang, Tikung, Yosowinangun,
Sukodono, Senduro, Padang, Pasirjambe.
Klakah, Ranuyoso, Gucualit, Randu
Agung, Jatiroto, Rowo Kangkung,
Kedungjajang.
Pasirian,
Candipuro,
Pronojiwo,
Tempeh, Kunir, Tempursari.
Jember, Jelbuk, Arjasa, Sukorambi.
Silo, Mayang, Pakusari, Kalasat,
Ledukombo, Mumbulsari, Sumberjambe,
Sukowono, Jombang.
Bangsalsari, Sumberbaru, Tanggul,
Semboro,
Umbulsari,
Kencong,
Gumukmas, Puger, Wuluhan, Balung,
Ambulu,
Jenggawah,
Tempurejo,
Rambipuji, Panti, Ajung.
Banyuwangi, Giri, Glagah, Kalipuro,
Wongsorejo.
Rogojampi, Kabat, Singoguruh, Srono,
Muncar,
Tegaldlimo,
Purwoharjo,
Bangorejo,
Pasanggaran,
Cluring,
Gambiran, Genteng, Sempu, Glenmore,
Kalibaru, Songgom.
Bondowoso, Tegalampel, Tenggarang,
Curahdami, Wringin, Pakem.
Klabang, Tapen, Prajekan, Cerme,
Wonosari, Sukasari, Tlogosari
Maesan, Tamanan, Pujer, Grujugan.
Situbondo, Mangaran, Panji, Kapongan,
Panarukan, Kendit.
Asembagus,
Arjasa,
Jangkar,
Banyuputih.
Besuki, Jatibanteng, Mlandingan, Suboh,
Sumber-malang,
Banyuglugur,
Bungatan.
Leces, Bantaran, Kuripan, Sumber,
Banyuanyar, Tegalsiwalan.
Kraksaan, Paiton, Kotaanyar, Pejarakan,
Krenjengan, Gending, Maron, Gading,
Pakuniran, Besuk, Krucil, Tiris
Tongas, Lumbang, Sukapura.
Gempol, Beji, Bangil, Rembang,
Pandaan, Prigen
Purwosari,
Purwodadi,
Sukorejo,
Wonorejo, Tutur, Puspo, Tosari

A- 11

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

217 5143

Grati

218 5151

Sidoarjo

219 5152

Waru

220 5153

Porong

221 5154

Krian

222 5161

Kabupaten /
Kotamadya
Code
Name

15

Sidoarjo

Mojosari

16

Mojokerto

223 5162
224 5171

Trowulan
Jombang

17

Jombang

225 5172

Mojoagung

226 5173

Diwek

227 5174

Tembelang

228 5181

Nganjuk

18

Nganjuk

229 5182

Kertosono

230 5191

Saradan

19

Madiun

231 5192
232 5201

Geger
Magetan

20

Magetan

233 5202

Maospati

234 5203
235 5211

Plaosan
Ngawi

21

Ngawi

236 5212

Padas

237 5213

Widodaren

238 5221

Bojonegoro

22

Bojonegoro

239 5222

Baureno

240 5223

Kalitidu

Carl Bro International a|s and Associates

Scope of Area
- Kecamatan
Grati, Rejoso, Nguling, Lumbang,
Winongan, Lekok
Sidoarjo, Candi
Waru, Gedangan, Sedati, Buduran,
Taman
Porong, Jabon, Tanggulangin, Tulangan
Krian, Balongbendo, Tarik, Wonoayu,
Prambon, Krembung, Sukodono
Mojosari, Bangsal, Pungging, Ngoro,
Kutorejo, Dlanggu, Gondang, Pacet,
Trawas
Trowulan, Jatirejo, Puri
Jombang, Megaluh, Peterongan
Mojoagung,
Sumobito,
Kesamben,
Mojowarno, Wonosalam,
Diwek, Ngoro, Gudo, Perak, Bareng,
Jogoroto, Bandar Kedungmulyo.
Tembelang, Ploso, Plandaan, Kabuh,
Kudu.
Nganjuk, Rejoso, Loceret, Berbek, Pace,
Bagor, Wilangan, Sawahan, Ngetos
Kertosono,
Baron,
Sukomoro,
Tanjunganom, Ngronggot, Prambon,
Jatikalen,
Patianrowo,
Lengkong,
Gondang Kulon, Ngluyu
Saradan, Mejayan, Pelangkenceng,
Balerejo, Gemarang, Wungu, Karee,
Wonoasri.
Geger, Dalopo, Kebonsari, Dagangan.
Magetan, Panekan, Parang, Lambeyan
Maospati, Sukomoro, Bendo, Takeran,
Karangrejo, Karangmojo, Kawedanan
Plaosan, Poncol
Ngawi, Paron, Geneng,
Padas, Karangjati, Kwadungan, Pangkur,
Bringin.
Widodaren, Kedunggalar, Mantingan,
Sine, Ngrambe, Jogorogo, Kendal, Pitu,
Karanganyar.
Bojonegoro, Parengan (Tuban), Sokosari
(Tuban), Bander, Bubulan, Trucuk.
Baureno, Kanor, Sumber rejo, Balen,
Kapas,
Kepohbaru,
Sugihwaras,
Kedungadem, Temayang, Margamulyo,
Sukosewu.
Kalitidu, Purwosari, Padangan, Malo,
Kasiman, Tambakrejo, Ngraho, Ngasem,

A- 12

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

Zone Name

Kabupaten /
Kotamadya
Code
Name

Scope of Area
- Kecamatan
Ngambon.

241 5231
242 5232

Tuban
Plumpang

243 5233

Bancar

244 5241

Lamongan

245 5242

Paciran

246 5243

Babat

247 5251

Gresik

248 5252

Sedayu

249 5253

Menganti

250 5261

Bangkalan

251 5262

Galis

252 5270

23

Tuban.

24

Lamongan

25

Gresik

26

Bangkalan

Sampang

27

Sampang

253 5280

Pamekasan

28

Pamekasan

254 5290

Sumenep

29

Sumenep

255
256
257
258

Kediri
Blitar
Malang
Probolinggo

71
72
73
74

Kediri
Blitar
Malang
Probolinggo

259 5750

Pasuruan

75

Pasuruan

260 5760

Mojokerto

76

Mojokerto

5710
5720
5730
5740

Carl Bro International a|s and Associates

Tuban, Semanding
Plumpang, Palang, Widang, Rengel.
Bancar,
Tambak
Boyo,
Jemu,
Merakurak, Montong, Kerek, Jatirogo,
Kenduruan, Bangelan, Senori, Singgahan
Lamongan, Deket, Tikung, Mantup,
Glagah, Karang Binangun, Kembang
Bahu
Paciran, Brondong, Laren, Solokuro.
Sukohadi, Karanggeneng, Sekaran,
Babat, Kedung Pring, Pucuk, Sugio,
Modo, Bluluk, Ngumbang, Sambeng,
Kalitengah,Turi, Sukorame, Maduran
Gresik, Kebomas, Manyar.
Sedayu, Dukun, Panceng, Ujung
Pangkah, Bungah
Menganti, Cerme, Benjeng, Balong
Panggang,
Duduk
Sampeyan,
Kedamean, Driyorejo, Wringinanom
Klampis, Sepulu, Tanjung Bumi,
Aresbaya, Kamal, Labang, Burneh,
Socah.
Galis, Tanah Merah, Blega, Tragah,
Kwanyar, Modung, Geger, Konang,
Kokop.
Sampang, Sreseh, Torjun, Camplong,
Omben,
Kedungdung,
Jrengik,
Tambelangan,
Banyuates, Robatal,
Sokobanah, Ketapang.
Tlanakan, Pademawu, Galis, Pamekasan,
Proppo,
Palengaan,
Pegantenan,
Larangan, Pakong, Waru, Batumarmar,
Kadur, Pasean.
Pragaan, Bluto, Saronggi, Lenteng,
Ganding, Guluk-guluk, Pasongsongan,
Ambunten, Rubaru, Sumenep, Kalianget,
Gapura, Manding, Dasuk, Batu Putih,
Batang-batang, Dungkek.
Kota Kediri.
Kota Blitar.
Kota Malang.
Kota Probolinggo.
Kota Pasuruan, Kraton, Pohjentrek,
Kejayan, Pasrepan, Gondangwetan
Kota Mojokerto, Gedeg, Kemlagi,
Sooko, Dawar Blandong, Jetis

A- 13

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix A:

Zone System Definition

Table: A.2: JARNS Road Network Model Study Area Zone System Definition
Seq.
No.

Hier.
No.

261 5770
262 5780

Zone Name

Madiun
Surabaya

263

6111

Bali

264

7111

Sumatra

Kabupaten /
Kotamadya
Code
Name
Madiun
77

78

Surabaya

Scope of Area
- Kecamatan
Kota Madiun, Jiwan, Sawahan.
Kota Surabaya

External Zone Representing Ferry Between Java & Bali

External Zone Representing Ferry Between Java & Sumatra

Carl Bro International a|s and Associates

A- 14

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

APPENDIX B:
TRAFFIC SURVEY DATA

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix B:

Traffic Survey Data

B1. TRAFFIC DATA COLLECTION

B1.1 Introduction
The main source of traffic data was the JARNS traffic surveys. A full decryption of surveys,
survey data collection program, data collected and its basic analyses are presented in Technical
Report No. 2 Traffic Surveys and Data Collection. A brief description of the surveys is given in
Section 3 of this report. This Appendix contains some key data items and summary tables, which
are useful to the readers of this report.
B1.2 Survey Locations and Program
Traffic surveys were conducted at 100 locations. Table B.1 provides a full description of survey
program, by sites by type and duration of surveys. For further details see Technical report No.2.
B1.3 Manual Classified Count (MCC) Surveys
MCC surveys were conducted at all (100) sites. Table B.2 gives a summary of 2-way total vehicle
counts by vehicle categories as used in the modelling process. Table B.3 describes the detail
vehicle types and defines the vehicle categories, and their compatibility to the vehicle type
definition used by other traffic databases.
Technical Report No.2 also includes a full database of JARNS traffic counts by time-period and
direction of travel by each of the 12 vehicle types. Therefore, readers interested in the breakdown
of the count data should refer to the JARNS MCC database.
B1.4 Other Traffic Data
Traffic count data was collected from a number of other studies and databases. These were:

Jasa Marga,

Automatic Traffic Count Data,

Pekalongan / Probolinggo SURIP Studies Data,

TSSS Data,

CAPEX Study Data

Data collected from these studies was used in JARNS where appropriate. The data obtained from
these studies is also included in Technical Report No.2 Traffic Surveys and Data Collection.
B1.5 Roadside Interview Surveys (RIS)
The complete set of RIS survey data has been put together in Microsoft Access database. This
database was used to build the observed trip matrices. A full description of this database is given
in Technical Report No.2 Traffic Surveys and Data Collection.

Carl Bro International a|s and Associates

B- 1

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix B:

Traffic Survey Data

Table B.1: List of JARNS Traffic Survey Sites and Description.

Site IRMS
Road
Func
Anode Bnode JARNS-Network Section & Description
No Link No.
Type
1
22001
A
2
27704 27705 20
Cilegon - Merak
2
22003
A
3
21933 21932 50
Sp. Bitung - Cikande
3
22008
A
3
21720 21520 10
Cikampek - Pamanukan
4
22009
A
2
21520 21430 10
Pamanukan - Janggaa
5
22012
K1
3
21420 27406 20
Karangampel - Cirebon
6
22013
A
4
27402 21130 30
Cirebon - Losari
7
22014
A
3
20310 20314 30
Cibinong - Bogor
8
22015
A
2
27104 20370 10
Bogor - Ciawi
9
22016
K1
2
20370 20390 10
Ciawi - Puncak (Cianjur)
10 22017
A
2
20520 20631 20
Selajambe - Padalarang
11 22020
A
2
21330 21313 30
Cileunyi - Sumedang
12 22021
A
2
21311 21320 10
Sumedang - Cijelag
13 22024
A
4
21120 27404 10
Palimanan - Cirebon
14 220311 K2
2
20213 20223 30
Cipanas - Rangkasbitung
15 22042
A
2
20401 27203 20
Cibadak - Sukabumi
16 22043
A
2
20370 20402 10
Ciawi - Cibadak
17 22045
A
2
27204 20512 10
Sukabumi - Cianjur
18 22049
A
3
20652 20720 30
Cileunyi - Nagreg
19 22056
A
2
20720 20721 10
Sasabengi - Malangbong
20 22062
A
2
20920 20921 10
Banjar - Cilapedang
21 22067
K2
2
21020 21011 30
Cirebon - Kuningan
22 22069
K2
2
21212 21210 10
Kadipaten - Majalengka
23 22071
K1
2
20340 27103 50
Parung - Bogor
24 22075
K2
2
20640 21530 20
Cagak - Bandung
25 22076
K2
2
21520 21511 10
Pamanukan- Subang
26 220782
A
2
20633 20634 30
Cisomang - Padalarang
27 22080
A
2
21721 21611 20
Cikampek - Sadang
28 22091
K2
2
27605 27604 30
Bekasi - Cileungsi
29 22092
K2
2
20320 20322 20
Cileungsi - Citeureup
30 220952 K2
4
20312 20320 10
Cibubur - Cileungsi
31 22101
K2
4
27502 21910 20
Tangerang - Serpong
32 22118
K2
2
21312 21314 20
Sumedang - Wado
33
34
35
36
37
38
39
40

24001
240041
24006
240071
24009
24010
240112
24017

A
A
A
A
A
A
A
A

4
2
4
5
2
3
2
2

21130
32713
37502
32510
32413
32213
32314
30813

32921
32714
32510
32511
32414
32210
32221
30851

Carl Bro International a|s and Associates

10
20
10
10
10
20
10
20

Losari - Pejagan
Tegal - Pemalang
Pekalongan - Batang
Batang - Banyuputih
Kendal - Semarang
Ungaran - Bawen
Ambarawa - Pringsurat
Magelang - Muntilan

Survey
Type
MCC
MCC
RIS
MCC
MCC
MCC
MCC
MCC
RIS
RIS
RIS
MCC
MCC
MCC
MCC
RIS
MCC
RIS
MCC
RIS
MCC
MCC
MCC
RIS
MCC
RIS
RIS
MCC
MCC
RIS
MCC
MCC
RIS
RIS
MCC
MCC
RIS
RIS
MCC
RIS

B- 2

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix B:

Traffic Survey Data

Table B.1: List of JARNS Traffic Survey Sites and Description.

Site IRMS
Road
Func
Anode Bnode JARNS-Network Section & Description
No Link No.
Type
41 24018
A
3
31014 31012 20
Prambanan - Klaten
42 24020
A
2
30920 30921 20
Boyolali - Kartosuro
43 24022
A
5
37205 37201 10
Surakarta - Palur
44 24027
K1
3
37602 32810 10
Tegal - Slawi
45 24028
K2
2
32920 32830 10
Ketanggungan - Prupuk
46 24029
K1
2
32813 32830 20
Slawi - Prupuk
47 240301 K2
2
32712 32731 20
Pemalang - Randu Dongkal
48 240312 K1
2
32931 30231 10
Bumiayu - Ajibarang
49 240331
A
2
30230 30234 10
Kr. Pucung - Wangon
50 240361
A
2
30230 30233 10
Wangon - Jeruklegi
51 240422
A
2
30224 30223 10
Sampang - Buntu
52 24045
K2
3
30215 30221 10
Purwokerto - Sokaraja
53 240502
A
2
30531 30223 10
Buntu - Gombong
54 240511 K1
2
30220 30331 10
Banyumas - Klampok
55 24057
A
2
30511 30520 10
Kebumen - Prembun
56 240662 K1
2
30720 32322 10
Kretek (KDT) - Parakan
57 24074
K1
2
32312 30820 10
Kranggan - Secang
58 24081
A
3
32140 32113 10
Semarang - Demak
59 24082
K2
2
32130 32131 10
Semarang - Godong
60 241032 K2
2
31530 31531 30
Purwodadi - Gundih
61 24105
K2
2
31311 31320 30
Karanganyar - Bts. Jatim
62 24106
K2
2
37202 31121 10
Surakarta - Sukoharjo
63 24118
K2
2
32530 32531 30
Blado - Bawang
64 241321 K2
2
32733 32732 10
Towel - Moga
65 241391 K3
2
Selo - Boyolali
66 24151
K2
2
32812 32814 40
Slawi - Randudongkal
67 24156
K2
2
32425 32213 10
Cangkiran - Ungaran
68
69
70
71
72
73
74
75
76
77
78
79
80
81

280051
280062
280092
28012
28016
28018
280192
280211
28026
28030
28031
28038
280411
280422

A
A
A
A
A
A
A
A
A
A
A
K2
K1
A

2
3
2
3
5
2
2
2
3
4
2
2
2
2

52121
51912
51715
57601
51524
51413
51330
57404
51410
50750
52330
52220
52421
52411

51910
51911
51720
51541
51510
58150
51331
51320
58160
57301
52313
52433
52520
52413

Carl Bro International a|s and Associates

10
20
10
20
20
20
10
20
20
10
20
30
10
10

Ngawi - Caruban
Caruban - Nganjuk
Jombang - Mj. Agung
Mirilip - Krian
Waru - Sidoarjo
Gempol - Pasuruan
Pasuruan - Pilang
Probolinggo - Baduan
Gempol - Pandaan
Purwodadi - Karanglo
Glandong - Tuban
Bojonegoro - Babat
Sadang - Gresik
Lamongan - Gresik

Survey
Type
RIS
RIS
RIS
MCC
MCC
MCC
MCC
RIS
RIS
MCC
MCC
MCC
RIS
RIS
MCC
MCC
RIS
RIS
MCC
MCC
MCC
MCC
MCC
MCC
MCC
MCC
MCC
MCC
RIS
RIS
MCC
MCC
MCC
RIS
MCC
MCC
RIS
RIS
RIS
MCC
RIS

B- 3

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix B:

Traffic Survey Data

Table B.1: List of JARNS Traffic Survey Sites and Description.

Site IRMS
Road
Func
Anode Bnode JARNS-Network Section & Description
No Link No.
Type
82 280461 K2
2
52431 51741 10
Babat - Ploso
83 280572 K2
2
57605 57604 10
Mirilip - Legundi
84 280681 K1
2
51920 51921 20
Madiun - Ponorogo
85 280691 K2
2
50221 50312 20
Dengok - Trenggalek
86 28070
K1
2
50430 50412 20
Trenggalek - Tulungagung
87 28072
K1
2
51820 50610 10
Kertosono - Kediri
88 280731 K1
2
57104 50630 10
Kediri - Ngantru
89 28076
K2
2
51711 51712 10
Jombang - Pulorejo
90 280791 K2
2
57103 50511 10
Kediri - Srengat
91 280812 K2
3
51614 51412 10
Mojosari - Gempol
92 280851 K2
2
50742 50743 40
Kandangan - Batu
93 28088
K2
2
57300 50720 10
Malang - Turen
94 28089
K1
2
57303 50712 10
Malang - Kepanjen
95 280911 K1
2
50730 50713 20
Blitar - Kepanjen
96 280982 K1
2
50830 50812 30
Turen - [Link]
97 281012 K1
2
51311 50821 10
Probolinggo - Grobogan
98 281041 K1
2
50820 50822 10
Wonorejo - Pondok Dalem
99 281601 K1
2
50131 50133 60
Pacitan - Panggul
100 28173
A
2
52511 52530 10
Bunder - Legundi

Survey
Type
MCC
MCC
RIS
MCC
RIS
RIS
MCC
RIS
MCC
MCC
MCC
MCC
MCC
MCC
RIS
MCC
MCC
MCC
MCC

Note: MCC Surveys were also conducted in parallel at all RIS sites.

Carl Bro International a|s and Associates

B- 4

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix B:

Traffic Survey Data

Table B.2: Summary of JARNS 24-Hour 2-Way Traffic Counts by Site

Site No. IRMS Link MC
Car
LGV MGV HGV Angkot Bus
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40

22001
22003
22008
22009
22012
22013
22014
22015
22016
22017
22020
22021
22024
220311
22042
22043
22045
22049
22056
22062
22067
22069
22071
22075
22076
220782
22080
22091
22092
220952
22101
22118
24001
240041
24006
240071
24009
24010
240112
24017

3,874 3,767
5,569 3,288
2,650 3,944
5,707 5,013
2,795 3,638
2,302 4,702
5,169 6,376
3,661 5,404
2,272 12,029
5,129 8,343
2,754 3,465
2,302 3,113
7,555 7,051
1,396
376
5,378 10,191
2,313 5,474
1,349 2,729
2,174 3,317
1,281 2,985
1,946 1,646
3,280 6,082
5,884 2,506
5,584 5,924
3,091 2,127
4,844 1,318
1,250 3,769
2,970 9,772
5,455 5,161
4,283 1,437
2,746 2,920
6,976 9,144
2,148 1,397
2,904 4,334
2,002 4,330
8,693 7,835
5,307 5,651
11,153 8,894
13,202 16,356
2,736 5,153
12,187 7,854

1,102
1,192
996
1,301
1,554
1,545
1,598
2,089
2,152
4,020
1,682
1,234
3,103
240
2,136
2,008
1,753
2,122
979
932
1,915
1,153
2,391
1,371
1,039
1,448
1,631
1,733
1,325
1,195
1,871
591
1,830
1,576
2,001
1,740
2,866
3,678
1,966
2,750

Carl Bro International a|s and Associates

1,092
4,482
4,486
4,647
2,995
4,384
1,793
1,739
1,788
3,581
2,713
3,049
6,534
549
1,991
3,645
2,742
4,046
2,231
1,623
3,199
1,225
2,744
802
890
4,047
6,064
3,348
3,100
3,822
4,683
585
5,378
4,631
4,853
4,458
5,561
5,022
3,527
4,305

448 3,095
1,419 2,547
3,131 1,230
2,477
820
1,415 2,512
1,605 1,974
130 8,054
51 15,049
73 6,309
441 1,261
382 3,910
426 2,953
1,803 7,885
55
910
341 9,710
1,774 6,134
607 4,442
256 7,094
220 1,003
642
649
374 2,677
26 3,200
252 5,801
22 2,487
50
798
1,659 2,437
2,583 3,746
1,570 3,930
1,854 6,215
436 1,818
371 4,675
14 1,200
3,018 4,755
3,079
523
2,361 2,829
2,032
557
3,030 1,009
1,210 2,585
639
379
585 3,539

1,275
662
3,078
2,815
1,770
2,715
1,102
630
1,104
1,388
589
442
3,777
15
482
603
558
1,433
786
457
894
133
793
100
69
696
1,655
2,612
1,977
411
401
106
3,547
2,645
1,682
2,040
3,099
3,419
648
1,840

AADT
14,653
19,159
19,515
22,780
16,679
19,227
24,222
28,623
25,727
24,163
15,495
13,519
37,708
3,541
30,229
21,951
14,180
20,442
9,485
7,895
18,421
14,127
23,489
10,000
9,008
15,306
28,421
23,809
20,191
13,348
28,121
6,041
25,766
18,786
30,254
21,785
35,612
45,472
15,048
33,060

B- 5

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix B:

Traffic Survey Data

Table B.2: Summary of JARNS 24-Hour 2-Way Traffic Counts by Site

Site No. IRMS Link MC
Car
LGV MGV HGV Angkot Bus
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81

24018
24020
24022
24027
24028
24029
240301
240312
240331
240361
240422
24045
240502
240511
24057
240662
24074
24081
24082
241032
24105
24106
24118
241321
241391
24151
24156
280051
280062
280092
28012
28016
28018
280192
280211
28026
28030
28031
28038
280411
280422

18,345
10,023
30,796
9,972
862
6,381
4,845
3,694
3,419
4,850
3,761
9,163
3,747
2,672
3,908
1,547
5,016
8,783
16,975
1,726
13,429
26,466
1,130
4,321
3,825
2,744
8,445
3,981
6,582
7,396
11,550
60,830
18,999
5,049
5,883
12,258
18,731
3,947
2,750
9,245
5,753

9,201
6,380
10,379
2,834
739
2,206
1,310
2,160
1,429
1,302
1,466
6,234
2,932
986
2,875
1,689
3,774
7,268
4,407
920
2,926
7,960
248
165
579
155
2,295
893
2,623
6,716
3,331
20,541
12,569
6,903
4,760
17,643
19,857
1,993
1,447
2,965
2,983

2,799
2,666
5,061
1,415
281
1,118
954
1,233
939
812
1,099
2,206
1,389
796
1,438
1,611
1,591
2,636
1,379
584
1,812
2,873
301
333
521
418
782
548
1,479
1,860
1,313
4,696
3,722
2,368
1,713
3,545
4,367
1,030
745
1,229
1,390

Carl Bro International a|s and Associates

3,611
2,364
2,610
2,611
1,719
3,099
1,383
1,952
2,140
1,369
3,529
1,731
2,204
1,119
2,634
1,048
1,801
3,984
2,074
1,179
443
1,372
391
190
777
897
155
2,499
1,551
3,812
2,183
5,020
5,149
5,671
2,321
6,743
3,912
1,903
1,585
1,607
2,373

711
746
384
291
148
328
49
686
229
267
363
240
928
167
664
50
113
2,429
318
184
8
138
6
5
479
891
2,555
466
1,693
3,240
2,375
810
2,116
859
2,743
376
255
2,724

2,391
2,602
5,424
4,075
714
342
1,241
570
791
248
1,828
1,379
1,103
568
1,362
437
193
378
2,304
381
1,718
1,147
97
686
762
601
1,913
1,092
2,499
823
911
7,859
2,475
1,560
1,452
2,889
4,007
989
1,279
541
1,097

AADT

1,306 38,364
1,773 26,554
3,822 58,476
1,509 22,707
252
4,715
2,200 15,674
962 10,744
1,534 11,829
624
9,571
337
9,185
694 12,740
2,969 23,922
1,285 13,588
822
7,130
1,313 14,194
616
6,998
1,161 13,649
2,820 28,298
1,230 28,687
573
5,547
206 20,542
1,614 41,570
256
2,423
5,701
27
6,496
52
4,867
29 13,619
1,035 10,527
1,230 16,855
2,226 25,388
2,214 21,968
116 100,755
2,154 48,308
2,198 26,124
639 17,578
660 45,854
1,362 53,095
607 13,212
405
8,587
232 16,074
897 17,217

B- 6

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix B:

Traffic Survey Data

Table B.2: Summary of JARNS 24-Hour 2-Way Traffic Counts by Site

Site No. IRMS Link MC
Car
LGV MGV HGV Angkot Bus
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100

280461
280572
280681
280691
28070
28072
280731
28076
280791
280812
280851
28088
28089
280911
280982
281012
281041
281601
28173

5,043
4,685
10,628
8,347
12,578
4,143
13,356
8,110
10,433
7,590
4,271
8,880
14,553
11,737
7,331
8,710
3,105
6,360
7,542

905
7,539
2,511
1,446
2,499
1,951
3,364
2,036
1,734
3,992
2,384
3,570
6,108
4,998
2,249
4,446
2,375
1,522
1,264

502
1,590
1,548
811
1,604
1,147
2,669
782
1,097
1,345
653
934
1,424
1,051
793
1,627
958
1,021
900

Carl Bro International a|s and Associates

888
8,893
972
1,088
2,069
1,334
2,040
808
1,072
3,573
531
2,066
833
1,350
1,737
2,787
956
1,504
2,564

326
3,637
132
70
132
834
947
237
281
1,685
290
309
249
577
210
729
265
801
375

692
211
2,081
863
1,414
1,234
2,800
1,252
830
952
1,304
1,485
2,796
1,848
884
1,254
88
1,034
539

68
662
967
190
417
547
1,095
384
67
414
438
472
568
494
88
1,115
699
375
10

AADT
8,424
27,217
18,839
12,815
20,713
11,190
26,271
13,609
15,514
19,551
9,871
17,716
26,531
22,055
13,292
20,668
8,446
12,617
13,194

B- 7

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix B:

Traffic Survey Data

Table B.3 JARNS Survey Vehicle Types and Compatibility with Other Data Sources
JARNS Vehicle Types

JARNS
Category

IHCM
Class

IRMS
Class

Vehicle Description

Toll
Class

Motorcycle

Motorcycle

Motorcycle

Motorcycle/ Scooter

N/A

Car

Car

Light Veh

Angkot

Angkot

Utility Veh

Light Goods Vehicles

LGV

Light Veh

Utility - freight

Small Bus

Medium HV

5a

Small Bus

Large Bus

5b

Large Bus

IIA

Medium HV

6a

Medium HV

6b

Large Truck

7a

Large Truck

Car/ Jeep/ Station

wagon
Small Public
Transport Veh

I
I

Bus
6
7
8
9
10
11
12

Large Bus
Medium Goods Vehicle
(2-Axles, 4-Wheels)
Medium Goods Vehicle
(2-Axles, 6-Wheels)
Truck Rigid 3-Axles
(HGV)
Truck Trailer (HGV)
Truck Semi-trailer
(HGV)
Non-motorised
Vehicles

Truck 2-Axles
(4-Wheels)
Truck 2-Axles
(6-Wheels)
Rigid Truck
3-Axle

IIA

7b

Truck + Trailer

IIB

Large Truck

7c

Tractor + Semi-trailer

IIB

N/A

Non-motorised
Vehicles

N/A

MGV

HGV

Not
Modelled

Carl Bro International a|s and Associates

IIB

B- 8

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

APPENDIX C:
MATRIX ESTIMATION

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

C1. INTRODUCTION
Section 4 of this report discribed the methodology and processes involved in estimating the base
year vehicle trip matrices. This Appendix presents the results compiled at each main stage of the
matrix estimation process.

C2. SUMMARY RESULTS

The results of the matrix estimation process are presented in two main forms:
1.

A summary of the full Origin/Destination trip matrices compressed to 10 sector to

sector tables. The sectors are defined to show major strategic trip patterns in Java,
and how the matrix estimation process estimated the un-observed trip movements.
Table C.1 below gives the sector definition in terms of zones included in each sector.
The sector boundaries are also shown in Figure C.6 at the end of this Appendix.
Table C.1: Number of Zones in the JARNS Study Area
Sector Number and Name

Sequential* Zone Numbers of Zones in Each Sector

1
Jabotabek & Banten
1-20, 61-72, 76-77, 264
2
Cirebon
38-60, 75
3
Bandung
21-37, 73-74
4
Tegal & Pekalongan
148-164, 169-170
5
Cilacap & Magelang
78-103, 165
6
Semarang
121-147, 167-168
7
Yogyakarta & Solo
104-120, 166, 171-176
8
Surabaya & Madura
218-221, 238-254, 262
9
Kediri & Madiun
177-192, 222-237, 255-256, 260-261
10 Malang & Eastern Java
193-217, 257-259, 263
NB: For the Definition of Hierarchical Zone Numbers, Please see Table A.1 in Appendix A.

2.

A figure showing how trip-length frequency distribution changes with each stage of
the matrix estimation, and how average trip length changes as more trips are
estimated and added to the observed trip matrix.

C2.1 O/D Sector to Sector Trip Tables

For each of the five vehicle categories (Car, LGV, MGV, HGV and Bus) following five trip
matrices were compressed to the 10x10 sectors.
(i)
(ii)
(iii)
(iv)
(v)

JARNS Observed,
TSSS matrix,
Prior matrix (formed from by combining the observed and the TSSS matrix),
Initial estimation of the matrix, (no intra-kabupaten inter-zonal trips), and
Final estimated trip matrix.

Carl Bro International a|s and Associates

C- 1

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

These compressed matrix tables are presented as Tables C1.1 to C1.5 for Car trips, C2.1 to C2.5
for LGV, C3.1 to C3.5 for MGV, C4.1 to C4.5 for HGV and C5.1 to C5.5 for the Bus in the
following section of this Appendix.
C2.2 Trip Length Frequency Distribution
At each stage of the matrix estimation process, the over all trip length frequency distribution was
computed and compared with the previous stage. Any changes were reviewed to confirm that the
major trip patterns in the study are not being distorted by the matrix estimation process. This
control of the trip length distribution in the matrix estimation process is crucial to the validity of
the overall trip patterns in the study area.
Figures C.1 to C.5 show the trip length frequency distribution for each of the five vehicle
categories, for which matrix estimation was carried out. Each figure within it shows five sets of
trip distribution patterns, corresponding to each of the five main stages of the matrix estimation
process given above in Section C2.1

Carl Bro International a|s and Associates

C- 2

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

Table C.1.1: JARNS RIS - Observed O/D Matrix ( Vehicles / Day ) - Cars

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

9,048

2,415

7,607

373

304

460

409

171

95

115

20,997

2,415

5,090

3,079

668

166

173

101

44

29

11,773

7,606

3,079

6,755

121

687

145

138

28

17

36

18,612

373

669

121

2,263

1,620

3,398

542

74

32

38

9,130

305

165

687

1,621

5,485

1,717

2,257

174

42

38

12,491

460

173

146

3,398

1,717 13,456

3,489

590

179

148

23,756

409

101

137

542

2,258

3,489 19,163

393

518

253

27,263

171

44

28

74

174

589

393

3,864

2,850

5,370

13,557

94

29

17

31

42

180

517

2,850

9,859

866

14,485

115

37

38

39

148

253

5,370

10

20,996 11,773 18,614

866 16,492

23,366

9,129 12,492 23,755 27,262 13,558 14,487 23,364

175,430

Table C.1.2: TSSS Synthesized O/D Matrix ( Vehicles / Day ) Cars

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

1
1

47,738

5,443 11,697

728

510

10

300

1,065

155

206

Total

100

67,942

5,443

8,368

2,281

496

175

76

141

28

24

17,038

11,698

2,281

5,078

280

301

115

212

17

20

43

20,045

728

496

280 10,457

2,416

4,388

1,075

49

64

58

20,011

510

175

302

2,415

4,815

3,030

4,949

144

90

63

16,493

300

76

115

4,388

3,030 21,100

7,499

311

304

107

37,230

1,064

141

211

1,075

4,949

7,500 42,267

486

1,357

200

59,250

155

29

17

49

144

6,266 16,278

66,436

206

24

20

64

90

305

10

100

42

59

63

106

310

486 42,702
1,357

6,266 13,883

3,024

25,239

3,024 20,733

40,613

67,942 17,040 20,043 20,011 16,493 37,230 59,252 66,436 25,238 40,612

370,297

201 16,278

Table C.1.3: JARNS Synthesized Prior O/D Matrix ( Vehicles / Day ) Cars

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

5,982

8,887

586

479

575

703

265

257

208

69,308

5,983 12,221

3,523

893

289

233

212

85

87

52

23,578

8,887

3,523

7,576

322

759

224

246

52

56

78

21,723

586

893

321

8,528

2,531

4,336

955

120

107

102

18,479

479

289

759

2,532

7,647

3,282

5,483

282

136

111

21,000

574

232

225

4,335

3,283 26,038

5,238

769

404

236

41,334

703

212

246

955

5,484

5,238 47,392

616

1,347

358

62,551

265

86

53

120

282

768

258

86

56

106

136

404

10

208

52

77

103

110

236

51,366

617 43,648

5,287 13,709

5,287 16,388

64,835

3,222

27,290

3,222 28,857

46,932

69,309 23,576 21,723 18,480 21,000 41,334 62,551 64,833 27,291 46,933

397,030

Carl Bro International a|s and Associates

1,347

358 13,709

C- 3

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

Table C.1.4: JARNS Matrix Estimation Initial Estimated O/D Matrix ( Vehicles / Day ) - Cars

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

46,052 11,433

11,434 11,809

8,714

263

293

457

555

170

10

179

Total

158

68,274

5,123

338

245

179

173

52

58

36

29,447

8,714

5,122

6,512

161

524

179

196

30

38

55

21,531

263

338

162

4,288

1,577

3,639

786

72

106

96

11,327

293

245

523

1,578

4,467

2,223

3,470

139

80

70

13,088

457

180

179

3,638

2,224 21,914

3,683

666

310

212

33,463

555

173

196

786

3,470

3,684 43,224

537

1,318

352

54,295

170

52

30

73

139

665

180

58

38

105

79

311

10

158

36

55

97

70

211

536 41,177
1,319

3,524 14,392

3,524 10,088

60,758

3,217

18,919

3,217 24,542

43,131

68,276 29,446 21,532 11,327 13,088 33,462 54,294 60,760 18,918 43,130

354,233

352 14,393

Table C.1.5: JARNS Matrix Estimation Final Estimated O/D Matrix ( Vehicles / Day ) - Cars

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

46,101 12,388

8,911

187

232

421

500

138

155

130

69,163

12,387 10,793

6,022

209

214

160

153

43

48

29

30,058

178

23

33

46

24,100

8,911

6,022

8,195

121

411

160

188

209

120

4,344

1,507

3,823

769

64

107

95

11,226

233

213

412

1,507

4,104

2,048

3,458

107

72

64

12,218

421

160

160

3,824

2,048 21,843

3,345

766

287

213

33,067

500

152

177

769

3,458

3,345 43,377

488

1,295

139

43

23

64

107

766

154

48

33

108

72

288

129

29

46

95

64

213

488 41,206
1,295

3,201

53,913
60,885

9,770

18,236

3,267

3,267 18,539

37,582

69,163 30,057 24,099 11,228 12,217 33,067 53,915 60,884 18,234 37,584

10

352 14,848

352

3,200 14,849

350,448

Figure C.1: Car - Matrix Estimation Trip Length Distribution

40%
35%

Average T rip Lengths

RIS=103, T SSS=101
Prior=96, Init-est=92
Final est imation=92

% of Trips

30%
25%
20%
15%
10%
5%
0%
0

0-50 50-100 100-150 150-200 200-250 250-300 300-350 350-400 400-450 450-500

500+

Distance (km)
RIS

TSSS

Carl Bro International a|s and Associates

Prior

Init-est

Final Est

C- 4

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

Table C.2.1: JARNS RIS - Observed O/D Matrix ( Vehicles / Day ) LGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

2,954

368

1,887

40

368

1,750

1,584

1,887

1,584

4,490

40

467

49

39

81

92

44

50

8
9
10

10

Total

92

50

47

18

21

5,516

467

81

44

25

12

4,340

49

494

36

30

8,587

1,395

868

1,138

156

26

13

4,158

494

867

3,406

514

790

63

37

34

6,325

36

1,139

514

3,917

1,057

238

70

34

7,141

24

30

156

790

1,056

8,021

131

232

88

10,578

47

13

26

63

239

131

2,046

951

1,145

4,668

18

37

70

232

951

5,532

257

7,115

20

13

35

33

88

1,145

256

4,886

6,482

5,515

39

4,340

8,586

4,158

6,327

7,139 10,580

4,666

7,115

6,484

64,910

Table C.2.2: TSSS Synthesized O/D Matrix ( Vehicles / Day ) - LGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

9,983

833

2,955

89

56

69

102

34

34

19

14,174

832

6,100

1,305

328

72

19

22

8,690

2,956

1,304

3,352

122

192

28

38

8,006

88

328

122

6,917

1,348

1,418

351

25

12

23

10,632

56

73

192

1,347

3,138

890

1,710

68

83

61

7,618

69

19

29

1,419

889

6,483

2,282

177

121

38

11,526

102

21

37

351

1,711

2,282

6,749

146

625

75

12,099

34

25

68

176

147

5,669

2,321

2,876

11,324

34

12

83

121

625

2,321

7,791

874

11,873

24

61

37

75

2,875

874

6,696

10,667

7,618 11,523 12,101 11,323 11,872 10,669

106,609

19

14,173

8,691

10

8,005 10,634

Table C.2.3: JARNS Synthesized Prior O/D Matrix ( Vehicles / Day ) - LGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

12,419

1,076

2,744

141

126

182

151

117

132

109

17,197

1,076

7,468

2,098

660

158

95

57

56

71

59

11,798

2,745

2,098

5,262

163

548

69

62

42

43

45

11,077

141

661

163

5,530

1,458

1,467

328

63

40

50

9,901

126

157

547

1,458

4,884

997

1,956

123

120

107

10,475

182

95

69

1,466

998

7,962

1,744

383

182

79

13,160

151

57

62

329

1,956

1,743

9,779

233

624

148

15,082

117

56

42

62

124

383

233

6,484

2,113

2,690

12,304

131

71

44

40

120

181

625

2,113

9,943

1,048

14,316

10

110

59

45

50

106

79

148

2,690

1,048

8,946

13,281

9,899 10,478 13,158 15,083 12,304 14,316 13,281

128,591

17,198 11,798 11,076

Carl Bro International a|s and Associates

C- 5

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

Table C.2.4: JARNS Matrix Estimation Initial Estimated O/D Matrix ( Vehicles / Day ) - LGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

6
83

123

137

9
64

10
91

Total

13,016

4,145

2,621

67

90

20,437

4,145

5,365

2,369

235

135

50

55

19

59

26

12,458

2,620

2,369

4,804

64

386

32

45

14

23

22

10,379

66

235

65

3,085

1,028

1,124

364

33

38

44

6,082

83

135

385

1,027

3,170

548

1,383

74

99

78

6,982

123

51

32

1,124

549

6,395

1,383

326

118

71

10,172

137

54

45

365

1,383

1,383

8,798

170

468

125

12,928

64

19

14

34

73

327

170

5,016

1,194

2,856

9,767

91

59

23

38

99

118

468

1,194

5,812

811

8,713

10

90

25

22

44

78

71

125

2,855

811

7,321

11,442

20,435 12,457 10,380

6,083

6,984 10,171 12,928

9,765

8,713 11,444

109,360

Table C.2.5: JARNS Matrix Estimation Final Estimated O/D Matrix ( Vehicles / Day ) - LGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

14,584

4,450

2,658

50

70

107

128

54

81

84

22,266

4,450

5,204

2,546

161

115

45

50

16

54

22

12,663

2,658

2,546

6,791

53

314

28

43

11

20

20

12,484

50

161

52

3,389

861

1,103

405

28

41

42

6,132

70

116

314

861

3,450

412

1,245

61

93

72

6,694

107

45

28

1,102

412

6,883

1,323

350

106

70

10,426

129

50

42

405

1,246

1,323

9,293

150

414

121

13,173

53

16

11

28

60

350

150

5,258

1,022

2,993

9,941

81

55

20

41

93

107

414

1,022

6,554

790

9,177

84

22

20

42

72

70

121

2,993

791

7,192

11,407

22,266 12,665 12,482

6,132

6,693 10,428 13,172

9,943

9,176 11,406

114,363

10

Figure C.2: LGV - Matrix Estimation Trip Length Distribution

45%
40%
Average T rip Lengths
RIS=89, T SSS=102
Prior=104, Init -est =102
Final est imation=98

% of Trips

35%
30%
25%
20%
15%
10%
5%
0%
0

0 -5 0 5 0 - 1 0 0 1 0 0 - 1 5 0 1 5 0 - 2 0 0 2 0 0 - 2 5 0 2 5 0 - 3 0 0 3 0 0 - 3 5 0 3 5 0 - 4 0 0 4 0 0 - 4 5 0 4 5 0 - 5 0 0

5 00 +

Distance (km)
RIS

TSSS

Carl Bro International a|s and Associates

Prior

Init-est

Final Est

C- 6

2
3

4
5

8
9
10

LEGEND :
Zone Boundary
Sector Boundary
Province Boundary

FIGURE C.6: TRIP MATRIX ESTIMATION,

STRATEGIC MOVEMENT SECTORS

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

Table C.3.1: JARNS RIS - Observed O/D Matrix ( Vehicles / Day ) MGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

4,492

2,021

4,179

324

272

561

2,021

2,720

2,323

755

184

4,179

2,323

5,291

146

666

323

756

146

1,686

272

184

666

560

285

124

279

97

460

156

10

286
13,028

10

Total

278

459

156

287

13,029

286

98

67

60

50

8,564

124

117

91

62

53

13,052

1,116

1,291

265

95

33

31

5,742

1,116

4,591

1,226

804

97

62

40

9,058

1,291

1,226

3,830

1,252

615

172

152

9,507

117

265

804

1,253

5,017

242

190

133

8,397

67

91

95

96

615

242

2,635

1,667

2,109

8,077

60

62

34

62

172

191

1,666

5,018

604

8,025

50

52

30

40

152

133

2,110

603

5,545

9,001

8,563 13,051

5,742

9,057

9,510

8,397

8,077

8,023

9,004

92,452

Table C.3.2: TSSS Synthesized O/D Matrix ( Vehicles / Day ) - MGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

27,437

4,610

6,451

657

352

437

568

372

324

230

41,438

4,610

5,320

2,505

560

188

127

113

39

53

40

13,555

6,450

2,506

5,103

333

287

99

136

56

78

53

15,101

657

560

333

7,092

1,761

1,806

530

79

68

58

12,944

352

187

288

1,761

4,437

2,192

1,674

85

139

48

11,163

437

127

99

1,805

2,192

6,673

2,884

485

297

112

15,111

567

114

135

530

1,675

2,883 18,225

180

511

124

24,944

372

39

57

79

84

486

180

7,604

4,086

4,277

17,264

324

52

78

68

139

297

511

4,086

7,065

1,761

14,381

230

40

54

57

49

111

124

4,277

1,761 10,654

17,357

41,436 13,555 15,103 12,942 11,164 15,111 24,945 17,263 14,382 17,357

183,258

10

Table C.3.3: JARNS Synthesized Prior O/D Matrix ( Vehicles / Day ) - MGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

32,227

5,316

5,471

633

415

774

498

601

397

399

46,731

5,316

7,429

3,400

1,020

300

373

204

111

143

110

18,406

5,471

3,401

5,880

371

731

196

183

133

141

99

16,606

633

1,019

371

5,688

1,809

1,784

507

158

112

96

12,177

415

301

731

1,810

6,218

2,379

1,914

165

202

97

14,232

774

372

195

1,784

2,379

7,974

2,233

933

440

247

17,331

498

204

184

506

1,915

2,232 19,831

331

544

219

26,464

601

111

134

158

164

8,152

3,564

4,051

18,199

933

331

397

143

140

112

202

440

543

3,564

8,666

2,037

16,244

10

399

110

100

96

97

247

219

4,051

2,036 13,365

20,720

46,731 18,406 16,606 12,178 14,230 17,332 26,463 18,199 16,245 20,720

207,110

Carl Bro International a|s and Associates

C- 7

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

Table C.3.4: JARNS Matrix Estimation Initial Estimated O/D Matrix ( Vehicles / Day ) - MGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

24,246

3,196

4,984

272

226

503

360

379

3,195

5,385

3,228

337

168

207

130

4,984

3,228

5,706

192

529

129

161

273

337

192

3,623

1,159

949

345

226

168

528

1,160

3,786

1,425

504

206

130

948

1,425

5,491

360

130

161

345

1,407

378

58

82

77

97

761

256

77

113

60

135

285

10

248

61

69

54

73

195

34,670 12,846 15,193

7,068

10

Total

256

248

34,670

58

77

60

12,845

81

114

69

15,193

78

60

53

7,069

1,406

98

134

73

9,004

1,627

760

285

195

11,571

1,626 16,092

260

390

179

20,950

260

6,450

2,094

4,256

14,513

389

2,095

4,164

1,378

8,952

179

4,255

1,379 10,621

17,134

9,005 11,571 20,949 14,514

8,953 17,132

151,901

Table C.3.5: JARNS Matrix Estimation Final Estimated O/D Matrix ( Vehicles / Day ) - MGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

25,338

4,110

4,464

180

165

429

310

300

211

194

35,701

4,110

6,847

4,157

209

141

175

112

47

64

48

15,910

4,464

4,156

7,137

149

452

110

152

63

101

55

16,839

180

209

149

3,950

1,077

838

314

61

51

45

6,874

165

141

452

1,077

3,958

1,277

1,482

82

130

70

8,834

428

176

110

838

1,276

6,044

1,550

796

250

181

11,649

310

112

152

314

1,482

1,550 16,256

238

362

169

20,945

300

46

63

61

82

796

238

6,370

1,794

4,701

14,451

212

64

101

51

130

249

362

1,794

4,705

1,299

8,967

194

48

56

44

70

181

170

4,701

1,299 10,463

17,226

35,701 15,909 16,841

6,873

8,833 11,649 20,946 14,452

8,967 17,225

157,396

10

Figure C.3: MGV - Matrix Estimation Trip Length Distribution

40%
35%

Average T rip Lengths

RIS=143, T SSS=183
Prior=131, Init -est =123
Final est imation=114

% of Trips

30%
25%
20%
15%
10%
5%
0%
0

0 -5 0 5 0 - 1 0 0 1 0 0 - 1 5 0 1 5 0 - 2 0 0 2 0 0 - 2 5 0 2 5 0 - 3 0 0 3 0 0 - 3 5 0 3 5 0 - 4 0 0 4 0 0 - 4 5 0 4 5 0 - 5 0 0

5 00 +

Distance (km)
RIS

TSSS

Carl Bro International a|s and Associates

Prior

Init-est

Final Est

C- 8

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

Table C.4.1: JARNS RIS - Observed O/D Matrix ( Vehicles / Day ) HGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

503

750

1,663

163

248

539

200

970

151

159

5,346

750

898

279

140

142

173

55

118

28

23

2,606

1,663

279

377

22

204

49

77

151

18

15

2,855

164

140

23

163

102

197

24

65

893

248

142

204

102

248

241

168

94

23

10

1,480

539

173

49

196

242

951

231

434

37

136

2,988

200

55

77

24

167

231

698

239

35

38

1,764

970

118

151

65

94

434

240

1,718

1,318

1,056

6,164

150

28

19

23

37

35

1,317

614

292

2,524

10

160

23

15

10

136

38

1,056

292

997

2,733

5,347

2,606

2,857

890

1,480

2,988

1,766

6,162

2,525

2,732

29,353

Table C.4.2: TSSS Synthesized O/D Matrix ( Vehicles / Day ) - HGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java

10

Total

2,521

1,685

2,439

341

371

400

443

727

320

137

9,384

1,685

360

245

110

137

78

54

53

27

28

2,777

2,438

246

370

44

92

40

89

71

24

3,423

342

110

44

597

196

278

55

43

20

1,691

371

137

92

196

209

417

313

65

50

12

1,862

400

78

40

278

417

1,628

501

172

63

73

3,650

444

54

89

55

313

501

673

179

91

31

2,430

727

53

71

43

65

172

178

4,737

3,233

1,660

10,939

320

28

23

19

50

63

92

3,232

863

952

5,642

31

1,660

952

1,180

4,089

2,429 10,939

5,643

4,088

45,887

Total

137

27

12

74

9,385

10

2,778

3,422

1,690

1,862

3,651

Table C.4.3: JARNS Synthesized Prior O/D Matrix ( Vehicles / Day ) - HGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

3,133

2,096

2,701

374

2,095

1,160

483

2,700

483

613

375

247

424

250

733

237

404

115

1,132

165

197

405

1,132

10

418

Total

424

733

266

11,682

247

250

237

115

165

90

77

4,919

77

256

89

147

198

58

55

4,676

78

640

268

300

67

104

51

58

2,188

256

268

418

566

390

151

84

54

2,861

89

301

566

2,204

567

541

113

212

5,563

147

67

390

567

1,044

342

131

85

3,292

104

152

541

342

5,541

2,999

1,928

13,101

418

91

58

50

84

112

131

2,999

1,328

1,193

6,464

10

266

77

55

59

53

212

85

1,928

1,194

2,065

5,994

11,680

4,921

4,677

2,187

2,861

5,561

3,293 13,101

6,466

5,993

60,740

Carl Bro International a|s and Associates

C- 9

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

Table C.4.4: JARNS Matrix Estimation Initial Estimated O/D Matrix ( Vehicles / Day ) - HGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

7,131

782

782

1,832

10

Total

1,832

87

221

328

200

301

229

152

11,263

827

476

44

68

76

41

39

46

50

2,449

476

1,015

19

87

38

66

46

40

43

3,662

87

44

18

96

74

72

16

20

12

21

460

221

69

87

74

232

324

153

72

55

64

1,351

327

76

38

72

324

1,375

198

230

50

387

3,077

200

41

66

16

152

198

213

78

33

46

1,043

301

39

46

20

72

230

78

3,666

1,033

846

6,331

229

46

40

12

55

50

33

1,033

519

787

2,804

10

152

50

44

20

64

387

45

846

787

616

3,011

11,262

2,450

3,662

460

1,349

3,078

1,043

6,331

2,804

3,012

35,451

Table C.4.5: JARNS Matrix Estimation Final Estimated O/D Matrix ( Vehicles / Day ) - HGV

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

7,630

1,005

1,777

55

161

291

155

187

185

111

11,557

1,004

1,927

498

24

54

63

32

28

42

41

3,713

1,777

498

1,474

25

62

35

46

23

32

33

4,005

55

24

25

304

71

68

69

14

38

16

684

161

54

62

72

348

278

123

53

56

68

1,275

292

63

35

68

277

1,252

200

172

44

508

2,911

154

32

46

68

123

200

332

45

26

39

1,065

187

28

23

15

53

171

46

3,420

1,022

1,037

6,002

185

42

32

38

56

45

26

1,022

985

986

3,417

110

41

33

16

69

507

38

1,037

985

743

3,579

11,555

10

3,714

4,005

685

1,274

2,910

1,067

6,001

3,415

3,582

38,208

Figure C.4: HGV - Matrix Estimation Trip Length Distribution

35%
30%

Average T rip Lengt hs

RIS=251, T SSS=201
Prior=221, Init -est=182
Final est imat ion=165

% of Trips

25%
20%
15%
10%
5%
0%
0

0 -5 0 5 0 -1 0 0 1 0 0 - 1 5 0 1 5 0 - 2 0 0 2 0 0 - 2 5 0 2 5 0 - 3 0 0 3 0 0 - 3 5 0 3 5 0 - 4 0 0 4 0 0 - 4 5 0 4 5 0 - 5 0 0

5 0 0+

Distance (km)
RIS

TSSS

Carl Bro International a|s and Associates

Prior

Init-est

Final Est

C- 10

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

Table C.5.1: JARNS RIS - Observed O/D Matrix ( Vehicles / Day ) Bus

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

270

1,087

1,479

797

456

300

336

103

81

98

5,008

1,087

225

359

21

45

137

1,885

1,479

359

1,854

47

353

109

39

12

4,260

797

21

47

1,555

405

704

23

16

27

3,595

456

45

353

405

3,438

273

389

16

5,379

300

137

109

704

273

3,544

1,195

421

51

6,734

336

39

23

389

1,195

4,299

427

104

66

6,884

103

12

16

16

421

427

769

1,131

1,299

4,194

81

27

104

1,131

1,058

359

2,769

10

98

51

66

1,299

359

748

2,631

5,008

1,885

4,260

3,595

5,379

6,734

6,884

4,194

2,769

2,631

43,339

Table C.5.2: TSSS Synthesized O/D Matrix ( Vehicles / Day ) - Bus

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java

10

Total

823

2,410

2,498

967

605

161

558

92

171

36

8,323

2,410

229

261

14

47

61

3,031

2,498

261

1,920

107

166

84

34

5,086

967

14

107

6,659

682

994

78

34

45

9,582

605

47

166

682

2,804

360

734

10

5,415

161

61

84

994

360

4,851

2,514

149

28

9,202

558

34

78

734

2,514

3,449

264

292

46

7,974

92

35

10

149

264

762

2,255

3,156

6,728

171

45

292

2,255

1,491

1,359

5,627

36

28

46

3,156

1,359

583

5,216

8,323

3,031

5,086

9,582

5,415

9,202

7,974

6,727

5,627

5,216

66,183

10

Total

Table C.5.3: JARNS Synthesized Prior O/D Matrix ( Vehicles / Day ) - Bus

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

1,348

3,038

3,038

421

3,087

513

1,481

57

973

6
7
8

3,087

10

Total

1,481

973

421

605

181

268

191

11,592

513

57

117

235

46

52

86

59

4,624

2,639

133

476

188

78

48

54

45

7,262

133

3,487

931

1,122

99

70

111

55

7,547

117

476

931

5,665

552

960

69

73

78

9,894

421

235

188

1,122

552

6,549

2,833

572

104

138

12,713

605

46

78

99

960

2,833

5,826

631

390

116

11,584

181

52

48

70

69

572

631

1,428

2,971

4,268

10,290

268

86

54

111

73

104

390

2,971

2,485

1,703

8,246

10

191

58

45

55

78

138

116

4,268

1,703

1,233

7,886

11,593

4,624

7,262

7,547

9,894 12,713 11,584 10,290

8,246

7,886

91,638

Carl Bro International a|s and Associates

C- 11

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix C:

Matrix Estimation

Table C.5.4: JARNS Matrix Estimation Initial Estimated O/D Matrix ( Vehicles / Day ) - Bus

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

1,367

840

840

1,908

1,908

657

1,537

730

730

1,317

657

13

30

341

36

263

83

408

8
9
10

10

Total

263

408

83

155

86

6,106

13

36

83

16

11

29

10

3,304

30

154

69

63

11

20

4,311

1,269

445

265

146

32

74

17

2,948

154

445

1,863

188

323

13

23

22

3,407

69

265

188

3,700

1,806

367

154

121

7,014

16

63

146

323

1,806

4,109

296

233

80

7,480

83

11

11

32

13

367

296

214

897

1,375

3,299

155

29

20

74

23

154

233

897

753

570

2,907

86

10

17

22

121

80

1,375

570

1,555

3,846

6,106

341

3,304

4,311

2,948

3,407

7,014

7,480

3,299

2,907

3,846

44,623

Table C.5.5: JARNS Matrix Estimation Final Estimated O/D Matrix ( Vehicles / Day ) - Bus

Origin / Destination
Jabotabek & Banten
Cirebon
Bandung
Tegal & Pekalongan
Cilacap & Magelang
Semarang
Yogyakarta & Solo
Surabaya & Madura
Kediri & Madiun
Malang & Eastern Java
Total

10

Total

2,548

802

1,589

482

233

214

332

58

115

57

6,431

802

2,645

712

10

32

78

17

27

4,341

1,589

712

1,169

24

125

50

67

18

3,766

482

10

24

1,212

437

471

120

29

67

19

2,872

233

32

125

437

1,911

162

393

12

25

23

3,353

214

78

50

471

162

3,657

1,650

320

146

113

6,860

332

17

67

120

393

1,650

4,600

256

235

81

7,752

58

29

12

320

256

268

820

1,227

3,007

115

26

18

67

25

146

235

820

936

573

2,963

57

19

23

113

81

1,227

573

2,005

4,113

6,431

4,340

3,766

2,872

3,353

6,860

7,751

3,007

2,963

4,113

45,457

10

Figure C.5: Bus - Matrix Estimation Trip Length Distribution

35%
30%

Average T rip Lengt hs

RIS=165, T SSS=151
Prior=173, Init -est=166
Final est imat ion=148

% of Trips

25%
20%
15%
10%
5%
0%
0

0 -5 0

5 0 -1

00

1 0 0-

1 50

1 5 0-

2 00

2 0 0-

2 50

2 5 0-

3 00

3 0 0-

3 50

3 5 0-

4 00

4 0 0-

4 50

4 5 0-

5 00

5 0 0+

Distance (km)
RIS

TSSS

Carl Bro International a|s and Associates

Prior

Init-est

Final Est

C- 12

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

APPENDIX D:
NETWORK DATABASE AND CODING

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix D:

Network Database and Coding

D1. INTRODUCTION
D1.1 Background
Section 5 of this report describes the methodology and the processes related to the development
of the JARNS network model and its components. This Appendix reviews the sources of data,
and explains how this data was interpreted, used, and coded in the JARNS network model.
Data for the network model were collected from a number of sources; these are described in
Section 2. Section 3 gives the description of the JARNS network model database, and its use in
the network simulation model. Section 4 describes the additional network data required for the
TRIPS modelling environment used for the Study.

D2. NETWORK DATABASE RESOURCES

D2.1 Overview
A great deal of work has been done in Indonesia over many years in the development,
implementation and refinement of a number of different Road Management Systems and
associated databases. The main database is the Indonesian Road Management System (IRMS)
database. In order to improve the quality and reliability of the IRMS data the Automated Road
Monitoring Study (ARMS) project commenced in 1999 and conducted a number of new surveys
using latest techniques, in a number of provinces in Indonesia, including all of Java. The data
from the ARMS surveys were made available to the JARNS project, too late for much of it to be
of use. The JARNS network database includes data items from both of these data sources. Only
appropriate and reliable data items, from either the IRMS database or the ARMS survey data
were selected. The overall data items available to JARNS from these data sources were:
1.
2.

The IRMS database contains road inventory and condition data for every link in the
JARNS network model.
The ARMS data collection programme collected a wide range of road data for all
roads in Java. This included digital image data of the whole of the JARNS network,
which was very useful in updating some of the key elements of the IRMS database.
It also provided detailed information on roadside land use and side friction, which
were essential in defining the speed-flow relationships.

This Section provides details of the major elements of these data sources used in building the
JARNS network model database. The main purpose is to emphasise the depth and value of the
road management system resources/databases that have been created in Indonesia, how these have
contributed to the JARNS network model, and how they may be used in other similar network
planning and modelling studies.
D2.2 IRMS Database and ARMS Survey Data
D2.2.1 Background

Carl Bro International a|s and Associates

D- 1

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix D:

Network Database and Coding

A comprehensive database for national and provincial roads was established in the mid 1980s
and populated for all Indonesia in 1987. This database is now the foundation for the Indonesian
Roads Management System (IRMS) database and contains data for all national and provincial
roads in Java.
The network is divided into links, normally at formal administrative boundaries, or at major
intersections. Data is stored for each of these links, and each link of the road network is identified
by a unique code identifying its location in Java. The database comprises numerous items related
to physical, geometric and other characteristics of road, and is stored in fields for each link. A
number of the database fields have been updated from the ARMS survey data. The most recent
database for Java now includes data for close to 11,365km of roads comprising of 726 links. This
survey data represents slightly smaller network of roads than the earlier IRMS database, as
ARMS surveys did not collect data for all the links in the IRMS database. This updated 11,365
km (726 Links) data from the ARMS surveys, and 74 km (20 IRMS Kota Links, not surveyed by
ARMS) of roads database was the basis for the JARNS study network database. IRMS links
included in the JARNS network model are shown in Figures D.1a, D.1b and D.1c for West Java,
Central Java and Yogyakarta, and East Java respectively.
D2.2.2 IRMS / ARMS Data Used by JARNS
The IRMS data supplied to JARNS was in the form of computer files. The key data items
received from the IRMS are discussed below. The network data received has been identified by a
unique combination of link number based on:
NOPROP
RUAS
SFFX

Province Number, a 2-digit code as used by BPS i.e. 22/24/26&28 for

West Java, Central java, Yogya and East Java respectively.
A 3-digit link number, e.g. 001, 025, 102
Suffix, up to 2-digit code some time blank or 1, 2, 3 or 11, 12, 21, etc.

Data fields were either one record per link or several records per sub-link depending upon the
type of information.
The IRMS network database items included in the JARNS network database were:
1.
2.
3.
4.
5.

LINK-ID:

As defined above, but the three fields (NPROP, RUAS and

SFFX) were concatenated to form a single filed called LINK-ID
FUNCT:
Link function class (A: Arterial roads, K1, K2 and K3 different
classes of collector roads)
WCARR:
Carriageway width, averaged over the link length and coded to
0.1m units
SH_WIDTHL: Left-side shoulder width average over the link length and coded
to 0.1m units Later updated from ARMS survey data.
SH_WIDTHL: Right-side shoulder width average over the link length and coded
to 0.1m units Later updated from ARMS survey data.

In addition to the above fields, data received also included: TERRAIN, EFFCAP, LANDUSE and
traffic counts by 12 vehicle categories. The data related to these fields were found to be not
accurate, and were not used.
The ARMS surveys provided three main data files:

Carl Bro International a|s and Associates

D- 2

Merak(
n
Cilegon 001.
l
Serang 002.

JAKARTA

Tangerang

026.1

n
003.

029.

>

004.

Bekasi

n091.
101.
026.2

l
Pandeglang 028.

027.1

027.2

l
Karawang.

071.

082.
030.

Pamanukan

006.

l031.1

Depok.

Rangkasbitung

092.

014.

Saketi

l
Indramayu.

009.
008.

007.

l
Cibinong

072.
073.

Cikampek

076.
010.

080.
077.

032.

079.

l
Purwakarta 078.1

l
Subang.

031.2

Bogor.

015.

016.

Palimanan 025.
(

023.

Cianjur.017.
l

034.

042.
040.

119.
037.3

l
Pel. Ratu

Sukabumi.
n

045.

075.

Padalarang(078.2

Sumedang.

n
Cirebon.
013.

069.

021.

l068.

l118.

Majalengka.

018.

044.1

039.

095.1

012.

l
Sumber 067.

022.

043.

024.

046.1

044.

019.
n
BANDUNG.

066.

020.

049.
065.

047.

l
Soreang

Kuningan

056.
050.

038.1

048.

l
Garut. 054.
046.2

051.

055.

064.

l
Tasikmalaya. 058. 060. l061.
Ciamis.

062.

038.2

052.

063.
059.

081.1

LEGEND :
Arterial
Collector-1
Collector-2
Collector-3
Toll Roads
Urban / Kota Links

> National Capital

n Province Capital

081.2

n Kotamadya
l Kabupaten Capital
(

Kabupaten City

FIGURE D.1a: JARNS NETWORK - IRMS LINKS

( JAKARTA & WEST JAVA )

[Link] IRMS Link No.

Kabupaten/Kotamadya Boundary

SCALE :

Province Boundary

10

25

50

100 km

3.

090.2

l
Jepara. 090.1
092.

024.

Rembang.

089.

012.

n
Cirebon.

l
umber 067.

085.2

013.

Kudus.
l

002.
l

001.
025.

066.

Brebes

Tegal

n
003.

005.

030.1

Pekalongan
n

120.

l
Slawi

Kuningan

Kajen

028.
151.

152.

008. l

Batang
007.1

118.

007.2
068.1

153.

029.
030.2

154.2

137.1
068.2

Purbalingga.

031.2

Purwokerto.
044.

042.2

066.2

047.

103.2

076.2

103.1

038.

050.1
l

Kebumen.

037.

Cilacap.

057.

002.2

Boyolali.

l
Mungkid. 017.

005.1

020.

Surakarta 023.
n

l
Sleman

l
080.

l
Klaten. 018.

159.
146.

006.
l

Wates.

011.

158.

004.

002.1

Karanganyar.
105.

006.2

n
Madiun

l
Sukoharjo

Purworejo

n
YOGYAKARTA

104.

106.

019.

079.
060.

005.2

n
Magelang.

064.1
058.1

037.2

015.2
l

050.2

Ngawi

Sragen.

076.1

064.2

036.

037.1

015.1

048.
049.

097.
098.2

148.2

014.

058.2

051.2

098.1

n
Salatiga.

074.

056.2

099.1

Purwodadi.

011.1

l
Temanggung.

066.1

055.

036.2

056.1

041.
133.

081.1

l
059.

Banjarnegara

054.

045.

043.

036.1 035.

063.

052.2

032.

033.1

Wonosobo.

099.2

101.2

l100.

l
Ungaran

130.
033.2

102.

010.

137.2

l 096.

Blora.

112.2

148.1

124.

062.

095.2

083.

156.

155.

113.1

061.

112.1

082.

113.2

132.2 132.1

031.1 135.

154.1

091.1

101.1

081.

SEMARANG.
n

009.

091.2

086.1

094.

095.1

086.2

085.1

Demak.

Kendal.

027.

l
.

Pemalang

004.2

l
Pati.

093.

Magetan

107.1

Wonogiri.

157.

l
117.

108.
064.

017.2

l
Bantul

l
Ponorogo

109.1

Wonosari.

069.2
030.

Trenggalek

109.2

127.
0

l160.1

Pacitan

LEGEND :
Arterial
Collector-1
Collector-2
Collector-3
Toll Roads
Urban / Kota Links

> National Capital

n Province Capital

n Kotamadya
l Kabupaten Capital
(

160.2

Kabupaten City

FIGURE D.1b: JARNS NETWORK - IRMS LINKS

( CENTRAL JAVA & YOGYAKARTA )

[Link] IRMS Link No.

Kabupaten/Kotamadya Boundary

SCALE :

Province Boundary

10

25

50

100 km

092.

Rembang. 094.
l
Pati.

095.1

093.

031.

091.1

Tuban

112.1

095.2

097.

098.1

001.

Ngawi

037.2
046.2
005.2

002.2

Jombang

005.1
002.1

Karanganyar.
Magetan

105.

oharjo

004.

062.

063.

003.

006.2

007.

072.

068.1

016.

l
Sidoarjo

052.2

017.

Mojosari

Gempol

(
081.2

076.

078.

018.

026.
027.

n
Pasuruan
019.1

086.

l
Ponorogo

021.2

n 021.1

028.

Situbondo
l

024.1

113.

101.1
085.1
084.

073.1

066.1

022.

Probolinggo

085.2

n
Kediri

108.

109.1

009.2

009.1

145.2

Waru

077.2

Wonogiri.
064.

162. 051.

012.
052.1

145.1

013.

077.1

l
Tarokan
l
117.

n 011.

047.

008.2

057.2

Mojokerto

006.1

Nganjuk

n
Madiun

057.1

l008.1

146.2

Pamekasan.

Sampang.

n
SURABAYA

173.

046.1

023.

Gresik

042.1
l
043.

l 024.

143.1
142.

042.2

Bojonegoro

037.1

Sragen.

l 038.

036.

098.2

104.

Lamongan

044.1

103.2

099.1

wodadi.

Sumenep.

Bangkalan.

039.1

099.2

101.2
100.
l

041.1

040.

l 096.

Blora.

112.2

041.2

041.3

029.2

114.

Bondowosol
112.

030.

069.2
073.2
109.2

Trenggalek

066.2

27.

n
Malang.

Srengat

074.
070.
l

Tulungagung

075.1

Kepanjen.

Blitar

091.2

l
065.

088.

099.

l
Lumajang.
090.2

160.1

111.2
104.1

104.2

Jember.
106.
107.

Pacitan

024.2

111.1

089.

160.2

110.

l 125.

109.

025.
129.
131.1

l
Banyuwangi.
134.
131.2

133.
132.

LEGEND :
Arterial
Collector-1
Collector-2
Collector-3
Toll Roads
Urban / Kota Links

> National Capital

n Province Capital

n Kotamadya
l Kabupaten Capital
(

Kabupaten City

FIGURE D.1c: JARNS NETWORK - IRMS LINKS

(EAST JAVA)

[Link] IRMS Link No.

Kabupaten/Kotamadya Boundary

SCALE :

Province Boundary

10

25

50

100 km

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix D:

Network Database and Coding

1.

Road Network Inventory data file ([Link]) This file contains one record per 1.0
km of road identified by the CHANGE field, starting a new record for each link in
the network.

2.

Road Geometry Data file ([Link]) This file contains one record per 10m section
of the road, identified by the CHANGE field, starting a new record for each link in
the network.

3.

Centreline locations data file ([Link]) This file contains one record per 10m of
the road identified by the CHANGE field, starting a new record for each link in the
network.

From these three ARMS survey data files, data on the following fields for each IRMS link were
extracted (source file shown in brackets) and combined to form one record per link.
1.
2.
3.
4.
5.

6.

CHAINAGE: used to identify start and end position of the IRMS links which
were divided for the JARNS network model. The total Chainage
also provide the confirmation of the coded link length. ([Link])
LENGTH:
Link Length in 10m units. ([Link])
SH_WIDTHL: Left-side shoulder width average over the link length and coded
to 0.1m units. ([Link])
SH_WIDTHL: Right-side shoulder width average over the link length and coded
to 0.1m units. ([Link])
TERRAIN:
Data in the [Link] file was found to be incorrect, and was
subsequently extracted form the [Link] file. The Criterion used
to define if a link is Flat/Rolling/Hilly (identified by one
character F/R/H) is detailed in Appendix E section E2.1.1.
FTCLASS
Road function class (A /K1 /K2/K3) updated the IRMS data field
FUNCT as described above.

The [Link] data file provided the road location mapping, used to show network maps using
the MapInfo Software. This file also provided the Co-ordinate data for the position of the ends
(Nodes) of each link or section in the JARNS network.
D2.2.3 ARMS Digital Image Data
ARMS surveys recorded the digital images of roads in Java. These images are available in 10m,
500m or 1.0 km intervals. JARNS used the 1.0km interval digital image data to augments/update
the IRMS/ARMS road inventory data fields, and transcribed data for additional road-side
characteristics for use in the determination of link Capacity and Free-flow speeds.
The data for each IRMS link in the JARNS network model were recorded for each 1.0 km subsection. Based on this information the IRMS links that extend into urban areas without formal
boundaries were sub-divided, and called JARNS sections of IRMS link. The following data items
were recorded for each 1.0 km section of the road:
Road Type Data
Typical road cross-section for each IRMS sub-section was noted under the following
categories:

Carl Bro International a|s and Associates

D- 3

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix D:

Network Database and Coding

Road Type Codes and Description

Code
2
3
4
5

Description
2 Lane 2-Way Undivided
4 Lane 2-Way Undivided
4 Lane 2-Way Divided
6 Lane 2-Way Divided

This road type information was further used to cross check and update the data on road
width taken from the IRMS database, where a road section has been upgraded. Wherever
the road cross-section changed between above types for road length greater than 1.0 km,
the IRMS link was divided into JARNS section, so that the coded road type would be
correct for as many sections of links as possible.
Road Side Friction
Roadside friction was recoded in the following five bands and coefficient values used were
1 to 5 as shown below:
1.
2.
3.
4.
5.

Very Low
Low
Medium
High
Very High

The information was coded for every 1.0km sub-section, and averaged over the length of
the link, or its subdivided section (JARNS Network Model Link) as appropriate.
Road Side Land Use
The roadside land use was taken as percentage of frontage development along the road as
described in the IHCM. It was not possible to determine the percentage of development
from the front facing digital images. The side-facing images were therefore used to
determine the percentage of frontage development on the right hand side of the road, and
this was assumed to be representative of both sides of the road. The values were recorded
in the following four bands:
Land Use
Intensity
Low
Medium
High
Urban Area

Land Use
Code
1
2
3
4

Estimated
Land Use
0-29%
30-59%
60-89%
>90%

LU
Factor
0.15
0.45
0.75
0.95

Wherever the land use exceeded 90%, the area was designated as an urban area without
formal boundary, and the IRMS link was subdivided into urban and inter-urban sections.
This process is described in detail in Section 5.2.3 and illustrated in Figure 5.4 of this
report.

Carl Bro International a|s and Associates

D- 4

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix D:

Network Database and Coding

D3. JARNS NETWORK DATABASE

D3.1 Overview
The above section described the data collection process from the IRMS Database and ARMS
survey data files. The JARNS network database comprises of a sub-set of the IRMS/ARMS
database of roads and other additional elements. The elements of the JARNS network database
are outlined in Section 5 and the network model database elements have been summarised in
Tables 5.1 and 5.2 in Section 5 of this report. Table D.1 below summarises the road network of
Java in the IRMS/ARMS data sources, and illustrates the amount of road network included in the
JARNS network. Other roads, which have been included in the JARNS database and described in
Section 5 are also summarised below
Table D.1: Description of JARNS Network by Function of Road
Link Description
By
Road Function

Roads in
IRMS/
ARMS

Arterial Roads (A)

Collector Road (K1)
Collector Road (K2)
Collector Road (K3)
Total
Sub-Totals Strategic &
Other Roads
IRMS Kota Links
(Not in ARMS Surveys)
Toll Roads
Toll Road Access Links
Urban Area Corridor
Madura Ferry Link

2,760
1,397
5,409
1,799
11,365

Total JARNS Network

JARNS Database Description of Road Types (km)

Strategic Strategic
Other
Other
InterInterUrban
Rural
Totals
Urban
Urban
Areas
Areas
1,946
593
168
2,707
1,190
147
58
1,395
3,112
318
114
572
4,116
56
10
30
96
6,304
1,068
340
602
8,314

N/A

7,372

942

8,314
74
439
67
182
7

6,304

1,068

340

602

9,083

D3.2 JARNS Network Database Description

The road network included in the JARNS network database has the attributes described below:
D3.2.1 Anode and Bnode
Every link in the JARNS network database is defined by the two nodes, representing the ends of
a link. These nodes are called Anode and Bnode. The node numbers were defined by the JARNS
project in such a way that the location of each Node/link is quite clear from the numbering
system. All nodes are numbered using a five digit code of the form: XYYZO
Where:

X represents the province in which the node is located. The codes used are:
1. Jakarta; 2. West Java; 3. Central Java; 4. Yogyakarta, and 5. East Java.

Carl Bro International a|s and Associates

D- 5

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix D:

Network Database and Coding

YY represents BPS Kabupaten or Kotamadya number in which the node lies. ( see
Appendix A for these codes ).
Z represents the last digit of the zone number in which the node lies, and
O represents the number of the node within a zone. The numbers used are from 0-9.
Number zero (0) is used for the node where the zone Centroid connector links in to
the network. The remaining nodes within the zone are numbered from 1 to maximum
of 9.

D3.2.2 Link Length

All links are given a unique link length, sometime referred to as Distance. The length is coded in
10m units, and was taken from the ARMS Chainage data (described above) in 1m units, and
rounded to the nearest 10m. All link lengths in the study reports refer to the sum of the link length
in different categories. e.g. Table D5.2 above.
It should be noted that the link length for a few links in the IRMS/ARMS data was found to be
inaccurate. These were corrected by referring and confirming from other sources. Therefore in
some cases the JARNS total link length may not exactly match the length taken from the
IRMS/ARMS sources.
Link lengths for roads not in the IRMS/ARMS data sources were taken from other sources:

Toll Roads from JASA MARGA Publications

Toll Road Access Links from MapInfo co-ordinates and adjusted for bendiness,
Urban Area Corridors Computed from MapInfo co-ordinates.

D3.2.3 Link Types

This code refers to the road type, of which 13 road types have been defined. The strategic
network only applies to road types 2 to 5. The road types for the strategic network were
determined from the Digital Image data as described above. For some of the links the road type
was corrected depending upon the carriageway width and the presence of the median. Table D.2
below describes the Link Types code in the JARNS network database:
Table D.2: JARNS Network Model Link Types
Link Type
2.
(2-Lane Single)
3.
(4-Lane Undivided)
4.
(4-Lane Divided)
5. (6-lane Divided)
Sub-Total (2,3,4,5)
6. Toll Roads

Carriageway
Width (m)
< 6.7m
>=6.7, < 12.0
Sub-total
>=10.0, <14.0
>=14.0
Sub-total
>=14.0, <18.0
>=18.0
Sub-total
>=21.0m
Dual-2

Carl Bro International a|s and Associates

Strategic Interurban (km)

3,701
2,312
6,013
162
32
194
25
14
39
58
6,304

Strategic Urban
(km)
55
515
574
211
79
290
57
17
74
130
1,068

Total
(km)
3,756
2,831
6,587
373
111
484
82
31
113
188
7,372
309

D- 6

Java Arterial Road Network Study (JARNS)

Link Type
7. Toll Roads
8. Toll Roads
Sub-Total (6,7,8)
9.
11.
12.
13.
14.
31.
All

Technical Report No. 10, Appendix D:

Network Database and Coding

Carriageway
Strategic InterWidth (m)
urban (km)
Dual-3
Dual-4
Total Toll Roads
Toll Road Access Links
IRMS Urban Links
IRMS Kota Links
Urban Area Corridor Links
Rural Area Non-Strategic Links
Ferry Link
Total Network

Strategic Urban
(km)

Total
(km)
106
24
439
67
340
74
182
602
7
9,083

D3.2.4 Jurisdiction Code

This Field is coded to represent the province in which the link lies. The codes used are taken from
the BPS coding system and are given in the table below:
Jurisdiction Code
21
22
24
26
28

Province
Jakarta
West Java
Central Java
Yogyakarta
East Java

This code was replaced in the TRIPS simulation network by different road classification codes
given in table below:
Jurisdiction Code
1
2
3
4
5
6
7
9

Road Classification for the

Estimation of Road Traffic Accident Costs
2-Lane Single Carriageway Width <5.0m
2-Lane Single Carriageway Width >=5.0m, <6.7m, and
Rural Area Roads (Road Type 14)
2-Lane Single Carriageway Width >=6.7m, <10.0m, and
IRMS/ARMS Urban and Kota Links (Road Types 11 and 12)
2-Lane Single Carriageway Width >=10.0m
4-Lane Undivided Carriageway (Road Type 3)
Divided Carriageways 4/6 Lane (Road Types 4 and 5), and
Urban Area Corridor Links (Road Types 13)
All Toll Roads and Access Links (Road Types 6, 7, 8 and 9)
Ferry Links and Centroid Connectors (Road Types 31 and 32)

D3.2.5 Capacity Index

Capacity Index refers to the speed-flow curve number. These are fully described in Appendix E to
this report.
D3.2.6 Speed or Time Flag

Carl Bro International a|s and Associates

D- 7

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix D:

Network Database and Coding

This code refers to the coding of Speed or Time (S or T). However, this field is compulsory for
the simulation model (TRIPS) network.
D3.2.7 Link Speeds
This field refers to the estimated (congested) link speeds in the base year. These values are only
used to estimate link travel times in the first iteration of the assignment model. However, this
field is compulsory for the simulation model (TRIPS) network.
D3.2.8 Link Capacity
This Field refers to the 1-way link capacity. The Link capacity calculation is fully described in
Appendix E to this report. This field is compulsory for the simulation model (TRIPS) network.
D3.2.9 Direction Code
This code refers to the directions of travel. i.e. A 2-way link is coded as 2, and 1-way link is
coded as 1. There are no one-way links in the JARNS network. However, this field is compulsory
for the simulation model (TRIPS) network.
D3.2.10 Road / Link Classification
This field identifies the type of link, implying its importance in the network. Table below lists
these Codes that have been used:
Link Classification
Code
I
D
U
K
X
C
T
A
F

Road Type and Data Source

Inter-urban Strategic Network Link
Urban Section of the Strategic Network Links, defined from
ARMS digital images
IRMS/ARMS Urban Area Links, usually identified with 2-digit
suffix code
IRMS Kota Links ( K Kota Filed in IRMS database)
Rural Area non-Strategic IRMS Links
Urban Area Corridors Defined by JARNS
All Toll Road Links
All Toll Road Access Links
Madura Ferry Link

D3.2.11 Road Side Development - Land Use Codes

These codes only apply to the strategic inter-urban and urban area links, otherwise the field is
blank. The data extraction process is described above in section D.2. Codes used are given in the
table below.
Land Use
Intensity
Low
Medium
High
Urban Area

Land Use
Code
1
2
3
4

Carl Bro International a|s and Associates

Estimated
Land Use
0-29%
30-59%
60-89%
>90%

LU
Factor
0.15
0.45
0.75
0.95

D- 8

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix D:

Network Database and Coding

D3.2.12 Roadside Friction Codes

These codes only apply to the strategic inter-urban and urban area links, otherwise the field is
blank. The data extraction process is described above in section D.2. Codes used are given below.
1.
2.
3.
4.
5.

Very Low
Low
Medium
High
Very High

D3.2.13 Carriageway Width

This field gives the average (over its length) carriageway width for all strategic inter-urban and
urban links only. For all other types of links, the field is left blank. The carriageway width was
taken from the IRMS database, as described above in Section D.2.
D3.2.14 Shoulder Width
This field gives the average (of both side shoulders over the length of the link) shoulder width for
all strategic inter-urban and urban links only. For all other types of links, this field is left blank.
The shoulder width was estimated from the ARMS Road Inventory data file, as described above
in Section D.2.
D3.2.15 Terrain Type
This field was derived from the ARMS Road Network Inventory data file. It defines an average
terrain type for strategic inter-urban and urban links only. For all other types of links, this field is
left blank. The codes used are given below, and are fully described in Section D.2 above and also
in Appendix E.

F Flat
R Rolling
H Hilly

D3.2.16 Link-ID
This is a unique link identification code. It is derived by concatenating the IRMS/ARMS Codes
of NOPROP (2-digit), RUAS (3-digit) and SFFX (blank, 1 or 2 or 3 or 11 or 12 or 21 etc). This
code provides a 2-way database linkage between the JARNS database and the IRMS/ARMs
database. This field represents a 5 to 7 digit code, depending upon the SFFX value. For the
JARNS database link, which has no corresponding IRMS/ARMS link this field is coded with the
province number and the RUAS and SFFX are given default codes of 999 and 99 respectively.
D3.2.17 JARNS LINK Section Number
This code is used to identify the IRMS/ARMS sub-divided links. This code takes the value of 10,
20 etc. For non-strategic links, this field value is coded as 00.
D3.2.18 Corridor Number

Carl Bro International a|s and Associates

D- 9

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix D:

Network Database and Coding

This code identifies a link or a number of links in a specific corridor. These corridor numbers
were used to identify future road improvement schemes and scenarios. They are defined by a 4digit code as shown and listed in Figure D.2.

D4. CODING OF SIMULATION NETWORK

In addition to the network links described in Section D2 and D3 above the simulation network
requires the following information.
1.
2.

Centroid Connector Links

Node Co-ordinates

The following section describes their use in the network model and the coding convention and the
values used for each of these items.
D4.1 Centroid Connector Links
These are notional links used to load the trips to/from a zone on to the remainder of the network.
The correct coding of these links is crucial, as the coded speeds and distance can affect the choice
of routing, and subsequently may influence the network evaluation process involving the network
travel times and distances. In the JARNS network model, there are few zones, which have more
than one Centroid Connector link. Therefore, the effect of having more than one Centroid
Connector links is avoided where possible. The coding of the Centroid connector lengths and
speeds have taken account of:

Size of the zone,

The network density within the zone,
The types of roads within the zone, and
The level of accessibility levels to the areas within the zone.

Bearing in mind the above criterion, all 262 Java zones were categorised in different ways. The
final grouping of zones is shown in Figures D.3a for the West Java and Jakarta; Figure D.3b for
Central Java and Yogyakarta; and Figure D.3c for east Java. These figures also show the coded
Centroid connector lengths. A fixed access speed of 15 km/h was coded for all Centroid
connector links. This represents a reasonable average speed for travel within the zone of the study
area by inter-urban traffic. The base year network model validation process showed that this
speed is reasonable, when modelled and observed journey times between key urban areas were
compared.
D4.2 Node Co-ordinates
The network simulation model requires network co-ordinates, so that the model inputs/outputs
could be viewed graphically. The co-ordinates are the full references in metres in the Universal
Transverse Mercator projection on the WGS84 datum, Zone 49 South. This is a standard grid
system for West Java around Jakarta. It gives negative Eastings at the western tip of Java, and
large Northings throughout. The simulation model (TRIPS) program MVGRAF can handle these,
and its output plots (in DXF format) can be readily converted into GIS layers matching other
mapping, or could be brought into MapInfo Software environment.

Carl Bro International a|s and Associates

D- 10

1,014
3,282

1,012

1,020
1,050

3,284

1,042

3,286
3,292

1,044
3,294

3,300

3,340
1,060

2,174
3,312

3,352

3,351

1,150

2,172

2,180
2,410
3,314
3,430
3,330
3,420

3,370
3,364

3,362

2,190

3,354
3,366

2,204 1,102
1,070

1,090

2,212

1,080

3,316

3,322

1,110
2,260 1,130
1,1202,250

1,104

2,214

3,324

Desc

No

CRD_2

Desc_2

No

CRD_3

Desc_3

No

CRD_4

CRD
1,012

Jakarta-Serang

25

2,180

Cilacap-Tegal

49

3,323

Pacitan-Trenggalek

73

4,210

1,014

Serang-Merak

26

2,190

Weleri-Parakan-Secang

50

3,324

Trenggalek-Kepanjen

74

4,220

Demak-Godong

1,020

Jakarta-Cikampek

27

2,202

Ponorogo-Solo

51

3,325

Kepanjen-Lumajang

75

4,230

3,325

Jepara-Demak/Kudus

1,030

Jakarta-Bogor

28

2,204

Solo-Purwodadi

52

3,326

Lumajang-Banyuwangi

76

4,240

Wirosari-Cepu

1,042

Cikampek-Purwakarta

29

2,212

Madiun-Ngawi/Caruban

53

3,330

Buntu-Parakan

77

4,250

1,044

Purwakarta-Padalarang-Bandung

30

2,214

Pacitan-Ponorogo-Madiun

54

3,340

Lohbener-Indramayu-Cirebon

78

4,310

1,050

Cikampek-Jatibarang-Cirebon

31

2,220

Tulungagung-Kediri-Kertosono

55

3,351

Semarang-Rembang

79

4,320

Trenggalek-Dengok

1,060

Cirebon-Semarang

32

2,230

Lumajang-Probolinggo

56

3,352

Rembang-Tuban

80

4,332

Malang-Kandangan

1,070

Semarang-Solo

33

2,240

Kepanjen-Malang

57

3,354

Tuban-Widang

81

4,334

3,326

Ngawi-Cepu

1,144

Yogyakarta-Bantul

1,142
2,230

2,240

3,323

No

3,380

2,270 3,390
1,106

2,220

2,202

3,318

3,356

Desc_4

Kandangan-Jombang

Ungaran-Sukoharjo

10

1,080

Yogyakarta-Solo

34

2,250

Purowejo-Pasuruan

58

3,356

Widang-Gresik

82

4,340

1,090

Yogya-Magelang-Bawen

35

2,260

Gempol-Mojokerto

59

3,362

Semarang-Purwodadi

83

5,000

1,102

Solo-Ngawi

36

2,270

Widang-Jombang

60

3,364

Purwodadi-Cepu

84

5,100

Jakarta - Merak Toll Road

13

1,104

Ngawi-Jombang

37

2,410

Ajibarang-Buntu

61

3,366

Cepu-Widang

85

5,200

Jakarta - Cikampek Toll Road

14

1,106

Jombang-Surabaya

38

3,282

Serang-Pandeglang

62

3,370

Gresik-Paciran-Tuban

86

5,300

1,110

Surabaya-Gempol

39

3,284

Pandeglang-Rangkas Bitung

63

3,380

Kamal-Sumenep

87

5,400

Jakarta Outer Ring Roads ( JORR )

16

1,120

Gempol-Malang

40

3,286

Rangkas Bitung- Bogor

64

3,390

Krian-Bunder

88

5,500

2,162 1,030

Jakarta - Ciawi Toll Road

15

2,164

Jakarta City Toll Road

12

1,020

Malang-Turen

11

1,012

Jakarta-Serpong Toll Road

17

1,130

Gempol-Probolinggo

41

3,292

Bogor-Sukabumi

65

3,420

Cilacap-Rawalo

89

5,600

1,142

Probolinggo-Situbondo

42

3,294

Sukabumi - Padalarang

66

3,430

Rawalo-Purwokerto

90

5,700

Palimanan - Kanci Toll Road

19

1,144

Situbondo-Banyuwangi

43

3,300

Cibadak-Pelabuhan Ratu

67

4,110

Pandeglang-Saketi

91

5,800

Jakarta-Cengkareng Toll Road

20

1,150

Bandung-Palimanan

44

3,312

Bandung-Ciamis

68

4,120

Cikande-Rangkas Bitung

92

6,000

Semarang City Toll Road

21

2,162

Bogor-Parung-Ciputat

45

3,314

Ciamis-Rawalo

69

4,130

Sadang-Subang

93

7,100

Surabaya - Gresik Toll Road

22

2,164

Parung-Serpong-Tangerang

46

3,316

Rawalo-Purworejo

70

4,142

Bandung-Subang

94

7,200

Surabaya - Gempol Toll Road

23

2,172

Ciamis-Kuningan

47

3,318

Purworejo-Yogyakarta

71

4,144

Subang-Pamanukan

24

2,174

Kuningan-Cirebon

48

3,322

Yogyakarta-Pacitan

72

4,150

Cileungsi - Jonggol - Selajambe

JABOTABEK AREA

Padalarang - Cileunyi Toll Road

18

3,286

FIGURE D.2: JARNS CORRIDORS

7,111

2,206

2,202
2,201

2,205

2,203
2,193

2,204

2,750

1,740

1,750
1,730

2,182
2,181

1,720
2,191

1,710
2,192

2,760

2,172

2,171

2,141

2,152

2,011
2,038
2,021

2,035

2,143
2,034
2,031

2,013

2,036

2,012

2,142

2,154

2,032
2,161

2,033

2,151

2,710
2,023

2,112

2,162

2,022

2,037 2,039

2,153

2,132

2,053
2,063

2,121
2,102

2,133
2,730

2,040

2,111
2,740

2,064

2,052
2,720

2,122

2,113

2,131
2,123

2,062

2,101

2,051

2,065

2,103

2,072

2,061
2,083

2,093

2,091

2,071

2,082
2,092
2,081

LEGEND :
3 km
5 km
7 km
10 km
15 km

Zone Boundary

XXXX

Province Boundary
Hierarchical Zone Number

FIGURE D.3a : CENTROID CONNECTOR LENGTH

( JAKARTA & WEST JAVA )

2,141

2,142
3,202
3,182
3,201
2,112
3,203

2,111

3,192

2,740

3,171

3,181

3,172

3,183

2,121
2,102

2,113

3,760
3,291

2,123

3,282

3,292

3,271

3,281

3,211

3,750
3,272

3,273

3,283

3,161

5,2
3,162

3,151

3,213

3,242

3,253

3,184

3,164

3,740

3,260

2,103

3,191

3,214

3,241

3,251
3,252

2,101

3,212

3,173

3,154

3,221

3,152

2,093
3,293
3,042

3,032
2,091
3,021
3,023
3,022

2,092

3,012
3,011

3,231

3,071

3,730
3,223
3,073

3,072

3,043

3,053

3,092

3,082
3,091

4,040

3,052

5,213

5,202

3,132

3,101
3,103

5,211

3,133
3,720
3,131
3,112

3,085
3,061

3,142

3,143

3,141

3,083
3,710
3,084
3,081

3,051
3,063

5,223

3,163

3,222

3,041

3,031
3,033

3,013

3,153

3,232

3,111

3,102

5,203

5,201

5,212

5,770
5,192

5,191

3,113
3,062

4,010

4,710
3,121

3,122

5,021

4,020
4,031

4,032
5,012

3,123
5,014
5,011

5,031

5,022

5,013
5,032

LEGEND :
3 km
5 km
7 km
10 km
15 km

Zone Boundary
Province Boundary

XXXX

Hierarchical Zone Number

FIGURE D.3b : CENTROID CONNECTOR LENGTH

( CENTRAL JAVA & YOGYAKARTA)

3,192

3,171

3,181

3,172

3,183

3,173

3,184
3,161

5,233

5,231

3,164

5,242

5,252

5,232

5,261

3,162

3,151
3,152

3,153

5,223

3,163

5,221

5,213

5,222

5,253

5,780

5,152

5,174

5,212

5,211

5,760
5,162

3,133

5,154

5,171
3,131

5,202

3,132
,111

5,203

5,201

5,770
5,192

5,182

5,191

5,172

5,181

5,161

5,151
5,153

5,141
5,750

5,173

5,121

5,061
5,062

5,142

5,710

3,122

5,021

5,075

5,074

5,063

5,123
5,122

5,132

5,111
5,112
5,113

5,082
5,071

5,720

5,011

5,133

5,730

5,041

5,014

5,740

5,031

5,022

5,012

5,143

5,131

5,043
3,123

5,290
5,280

5,241

3,142
3,141

3,121

5,270

5,251
5,243

3,143

5,262

5,013
5,032

5,081

5,101

5,091

5,052
5,051
5,072

5,042

5,092

5,083
5,073

5,093

5,102

LEGEND :
3 km
5 km
7 km
10 km
15 km

Zone Boundary
Province Boundary

XXXX

Hierarchical Zone Number

FIGURE D.3c : CENTROID CONNECTOR LENGTH

( EAST JAVA )

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix D:

Network Database and Coding

Under this co-ordinate system, the approximate references for key locations are as follows:
Location
Merak
Jakarta
Yogyakarta
Surabaya
BanyuwangI

Easting
-60,000m
37,000m
430,000m
693,000m
873,000m

Northing
9,340,000m
9,316,000m
9,140,000m
9,200,000m
9,090,000m

Because the numbering of zones is not continuous, and there are many gaps in the node
numbering sequence, TRIPS hierarchical node numbering system is used as defined in Section
D.2.3.1 above.
Node co-ordinates were updated to match centreline data from ARMS surveys in [Link] files.
Intermediate nodes located, where an IRMS link is subdivided are located by chainages taken
from the digital images survey data.

Carl Bro International a|s and Associates

D- 11

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

APPENDIX E:
ROAD CAPACITY AND
SPEED FLOW RELATIONSHIPS

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

E1. INTRODUCTION
E1.1 Background
The calculation of link capacity and the relationship between the volume of traffic and traffic
speed is crucial to the JARNS network model. It is these relationships that identify the operational
characteristics of the network from which performance measures used for economic and
operational evaluations are derived.
The Indonesia Highway Capacity Manual (IHCM) defines these relationships for roads in
Indonesia. A review of these relationships was undertaken, and it was found that they are
relatively complex and demanding of data. For inter-urban strategic network analysis, it is
important to capture the essential structure of the relationships, but reduce the detail to facilitate
their use in practical modelling, and the TRIPS network system requirements.
Building on the work done in IRMS and the development of the Strategic Expenditure Planning
Model (SEPM), the IHCM relationships were reviewed and streamlined, to produce coherent,
consistent and manageable Speed Flow relationships for JARNS.
This Appendix describes the relationships developed for the JARNS network model. The
Appendix has three main sections. The first section describes the calculation of road capacities for
the strategic network links, and documents the data sources of the influencing variables. Section 3
details the calculation of Free Flow speeds for all links in the strategic network, and the source of
influencing factors. The final section then documents the final speed / flow relations used in the
JARNS network model.
E1.2 Summary of Input Data
Table E.1 below lists the data items and their sources, used in the calculation of speed-flow
relationships.
Table E.1: Summary of Data Items Used for Speed Flow Relationships
Data Item
Required for
Speed-Flow
Relationships
Road Type

IRMS/ARMS
Database Field

JARNS
Database

ARMS Digital Images

Link-Type

TERRAIN

Terrain

WCARR

Carr-Width

SH_WIDTHL &
SH-WIDTHR

Shld-Width

Roadside Land Use

ARMS Digital Images

Land Use

Roadside Friction

ARMS Digital Images

Rd-Friction

Terrain Type
Carriageway Width
Shoulder Widths

Carl Bro International a|s and Associates

Description and
(Data Source)
Single or Dual and Number of Lanes
Recorded by JARNS
(ARMS Digital Image Database)
Flat, Rolling, or Hilly
(IRMS Road Geometry Data)
2-Way Width
(IRMS Database)
Averaged of Both Sides
(IRMS Database)
Roadside Land Use (%)
Recorded by JARNS
(ARMS Digital Image Database)
Side Friction (Very low, to Very high,
Recorded by JARNS
(ARMS Digital Image Database)

E- 1

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

E2. CALCULATION OF ROAD CAPACITY

E2.1 Strategic Network Road Capacity
Road capacity for each link of the strategic network was calculated. All values were computed for
2-way capacity. The network simulation model requires link capacity to be input as 1-way
capacity.
Link capacity is function of a number of variables. The simplified form adopted from the IHCM
is:
Capacity = FacD * Co * F_CW * F_SF&SW
Where:
1. Capacity is the 2-way link Capacity in PCU/hr
2. FacD is the adjustment factor for the effect of directional traffic split.
= 0.97 Default value taken from IHCM.
3. Co is the Base Capacity (Section E2.1.1 below)
4. F_CW is the effect of Carriageway Width on the Base Capacity (Section
E2.1.2)
5. F_SF&SW is the adjustment factor applied to the Base Capacity to reflect the
effects of road Side Friction and Shoulder Width (Section E2.1.3)
The computation of each of the above variables is defined in following sections. These values are
computed for each link in the strategic network, and then combined to give the final 2-way
capacity. One-way link capacity was then simply taken as half of the two-way capacity and
rounded to the nearest 100. This capacity value was the directly coded into the network model.
In the forecast year scenario, if the link is upgraded, link capacity is computed as function of the
new link attributes, and the new capacity value coded into the network model database.
E2.1.1 Base Capacity
Base capacity depends upon the type of road and the terrain it goes through. The base capacity
values for a road section (link) were taken from the IHCM, and these are specified in the table
below. The road type for each link of the strategic network was initially taken from the IRMS
database. Where the IRMS links were subdivided into two or more JARNS network model links,
the road type was reconfirmed from the digital image database. The data on the type of terrain
was taken from the IRMS database (details below). Therefore, for each link of the network road
type and terrain were recorded in the JARNS road network database, for the derivation of the
Base Capacity (Co).
Co = Base Capacity (PCU/hr 2-way) as a function of Road Type and Terrain

Carl Bro International a|s and Associates

E- 2

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

Table E.2: Base Road Capacity by Road Type and Terrain

Road Type

Terrain
F
R
H

3
4-Lane Single

Road Type

F
R
H

2
2-Lane Single

Co
3,100
3,000
2,900

4
4-Lane
(Divided)

Terrain
F
R
H

Co
7,600
7,400
7,200

6,800
6,600
6,400

5
6-Lane
(Divided)

F
R
H

11,400
11,100
10,800

The type of terrain for each IRMS link was defined according to the table below. The same
terrain type was adopted for each section of the IRMS link.
Code

Description

F
R
H

Flat
Rolling
Hilly

Vertical Alignment
Range (m/km)
<10
10-30
>30

Horizontal Alignment
Range (deg/km)
<60
60-150
>150

E2.1.2 Adjustment Factor for Carriageway Width (F_CW)

The base capacity determined from above is adjusted for the effects of variable carriageway
widths within the same road type. This is estimated from the equation below:
F_CW = W1 + W2 * XCW + W3 * XCW2
Where:
F_CW is the base capacity adjustment factor based on:
(i)

XCW is the effective road width and is estimated from the Carriageway Width (CW)
as detailed below:
CW Total Two Way Carriageway Width (m)
Total two-way width of a link was taken as the average width along the
entire link. IRMS database provided the road width for every 1.0 km
section of the road for all IRMS links. Average value for the entire link or
section of a link was estimated and used. The maximum width to which the
road capacity increases were adopted from IHCM for each road type and
are detailed below:
For Road Type 2
For Road Type 3 and 4

XCW = Minimum (CW or 17.5)

For Road Type 5

(ii)

XCW = Minimum (CW or 12)

XCW = Minimum (CW or 26.5)

W1, W2, and W3 are model parameters taken from the IHCM, and are set out in the
table below:

Carl Bro International a|s and Associates

E- 3

Java Arterial Road Network Study (JARNS)

Road Type
2
3&4
5

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships
W1
-0.36
-0.47
-0.47

W2
0.274
0.175
0.117

W3
-0.0115
-0.0050
-0.0022

E2.1.3 Factor for Side Friction and Shoulder Width (F_SF&SW)

Link capacity is affected by the amount of side friction and shoulder width. The overall effect is
estimated from the model given below:
F_SF&SW = P1 P2 * SFC * Maximum [(3 SH_WIDTH) or 0]
Where:
(i)
P1 and P2 are the model parameters with values below:
1. P1=1.020 for all road types (2, 3, 4 & 5)
2. P2 = 0.018 for road types 2&3 (undivided roads), and
3. P2 = 0.012 for road types 4&5 (divided carriageway roads).
(ii)

Road Side Friction Coefficient (SFC) was recorded from the ARMS digital images
for every 1.0 km section of the road, and an average value was adopted for each
link in the network model. The roadside friction was recorded in 5 categories and
coefficient values adopted are given in the table below:
Road Side Friction
as Recorded
Very Low
Low
Medium
High
Very High

(iii)

Coefficient Value
(SFC)
1
2
3
4
5

Road shoulder width values for both Left and Right shoulders were available from
the ARMS database by each 100 m section of the road. Average values for the
entire length of the link were estimated for each link in the network model. The
value input to the model were:
SH_WIDTH = Shoulder Width (m) is the average Left and Right shoulder widths.
i.e. SH_WIDTH = (Left Shoulder Width + Right Shoulder Width)/2.

E2.2 Road Capacity for Remainder of the Network

As discussed in Section 6 of this report the network has a number of components, including:

inter-urban (strategic) links,

urban links

toll roads and access links

Carl Bro International a|s and Associates

E- 4

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

Each type of link has different speed-flow and capacity characteristics, and requirement for fully
representing these characteristics are also different.
The capacity measures described above relate to the inter-urban network, which is the main focus
of JARNS. While it is important that the network adequately represent the effect of urban links on
traffic flow in Java, there is no particular interest in quantifying this effect for project evaluation
purposes.
Consequently, for other components of the network, detailed capacity and speed-flow
relationships were not developed, and simplifying assumptions were used instead. These are
summarised below.
E2.2.1 Urban Area Links and Corridor Capacities
The link capacities adopted for all urban area links and notional urban area corridors were taken
as 50% of the 4-lane divided carriageway capacity (6,600 PCU/hr 2-way). Therefore, the coded
capacity for these links was taken as 3,300 PCU/hr 2-way and 1,700 PCU/hr 1-way (value
rounded to the nearest 100). This reflects the fact that most urban areas have divided roads (or
more than one through route), and half of the capacity is available for through traffic. See Table
E.3
E2.2.2 Toll Roads and Toll Road Access Link Capacities
Capacity values for toll roads depend upon the number of lanes, and were taken directly from the
IHCM and were coded in the network database. Toll roads access links, are notional links in the
network to provide connectivity between the toll roads and the remainder of the network. Their
capacity was taken as 2/3 of the 4-lane toll road links. The reason for adopting capacity values
lower than the toll road were:

These sections of the road are not designed to the same standard as the toll roads,

There is considerable roadside friction from commercial activity along these sections.

Table E.3 below details the road type and the coded link capacities.
E2.2.3 Non-Strategic Link Capacities
These are rural area roads, which were included in the network for accessibility to low density
rural areas from the strategic network. Their capacity was estimated to be about 85% of the
average base capacity of a single 2-lane road of 3,000 PCU/hr 2-way, equal to 1,300 PCU/hr 1way. (See Table E.3)

Carl Bro International a|s and Associates

E- 5

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

Table E.3: Coded Capacities for Non-Strategic Network Links

Capacity (PCU/hr) 1-Way

Road Type
(Link Type)

Description

6
7
8

4-Lane Toll Road

6-Lane Toll Road
8-Lane Toll Road

3,800
5,700
7,600

3,800
5,700
7,600

9
9

Toll Road Access Links (1)

Toll Road Access Links (2)

2,500
-

3,800
2,500

50%
-

11
12
13
14

IRMS Urban Links

IRMS Kota Links
Urban Corridors
Rural Area Links

1,700
1,700
1,700
1,300

3,300
3,300
3,300
1,700

100%
100%
100%
30%

2000
& 2010

2020

%
Increase

(1) Existing Toll Road Access Links

(2) Future Toll Road Access Links
Table E.3 above also gives the link capacities adopted for future year urban networks. While there
is no need for JARNS to formally, model urban traffic flows, it is also important to ensure that
overall network capacity is not significantly reduced by the presence of congestion on urban
links. It has been assumed that additional urban capacity will be provided as needed for growth in
overall traffic.

E3. FREE FLOW SPEED

The second most important element of the speed/ flow relationship is the link Free Flow Speed
(FFS), which determines the maximum safe driving speed possible. Estimates of FFS for the
strategic road network were derived from simplified versions of the IHCM models. This section
details the relationships used for the JARNS network model.
E3.1 Strategic Network Free Flow Speed (FFS)
Free flow speed is function of: road type, terrain type, carriageway width, shoulder width,
roadside friction, and the roadside land use. The model took the form of:
FFS = (F_Vo + F_CW) * F_SF&SW *F_LU
Where:
(i)

FFS = Free Flow Speed in km/h;

(ii)

F_Vo is the Base FFS based on Road Type and Terrain. (Section E3.1.1)

(iii)

F_CW is the effect of Carriageway Width on the Base FFS (Section E3.1.2)

(iv)

F_SF&SW is the adjustment factor applied to the Base FFS to reflect the effects of
road Side Friction and Shoulder Width (Section E3.1.3)

(v)

F_LU is the adjustment factor to reflect the effects of roadside Land Use (LU).
(Section E3.1.4)

Carl Bro International a|s and Associates

E- 6

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

E3.1.1 Base Free Flow Speed (F_Vo)

The base FFS speed was taken from the Table E.4 below, depending upon the Road Type and
Terrain of the link.
F_V0 = Base Free Flow Speed (km/hr) as a function of Road Type and Terrain

Table E.4: Base Free Flow Speed (F_V0) by Road Type and Terrain
Road Type
2
2-Lane Single
3
4-Lane Single

Terrain
F
R
H

F_V0
68
61
55

Road Type
4
4-Lane
(Divided)

Terrain
F
R
H

F_V0
78
68
60

F
R
H

74
66
58

5
6-Lane
(Divided)

F
R
H

83
71
62

The type of terrain for each IRMS link was previously defined.
E3.1.2 Adjustment Factor for Carriageway Width (F_CW)
The base FFS determined above is adjusted for the effects of variable carriageway widths within
the same road type. The effect is estimated for each road type using the following formulations:
Road Type = 2 and CW < 5;

F_CW = 9.15 * (CW 5) 9.244

Road Type = 2 and CW >= 5;

F_CW = 4.91 * [1 (15 / (CW 2.5)1.8) ]

Road Type = 3 or 4;

F_CW = Minimum [1.5 or 6 * (CW/4 3.5)]

Road Type = 5;

F_CW = Minimum [1.5 or 6 * (CW/6 3.5)]

Where:
(i) CW is the effective road width and is estimated from the Carriageway Width
as detailed below:
CW:= Total Two Way Carriageway Width (m)
Total two-way width of a link was taken as the average width along the entire link. The IRMS
database gives road width for every 1.0 km section of all IRMS links.
E3.1.3 Factor for Side Friction and Shoulder Width (F_SF&SW)
Link FFS is affected by the amount of side friction, and shoulder width. The overall effect is
estimated from the model below:
F_SF&SW = F1 F2 * SFC * Maximum [ (4-SH_WIDTH) or 0 ]
Where:

Carl Bro International a|s and Associates

E- 7

Java Arterial Road Network Study (JARNS)

(i)

F1 and F2 are model parameters; dependent on the road type and are given in the
table below:
Road Type
2
3
4 or 5

(ii)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

F1
1.05
1.03
1.03

F2
0.017
0.012
0.010

The determination of the Road Side Friction Coefficient (SFC) and the Shoulder
Width (SH_WIDTH) has been described above in Section E2.1.3.

E3.1.4 Adjustment Factor for Road Side Land Use (F_LU)

The road FFS is affected by the amount of roadside land use. This effect is estimated according to
the road type and the amount of land use adjacent to the road using the following formulations:
F_LU = 1.0 F3 * LU
Where:
(i)

(ii)

F3 = model parameter, with values below:

F3 = 0.060 for Road Type 2.
F3 = 0.055 for Road Type 3
F3 = 0.050 for Road Types 4 & 5 (Divided Roads).
LU = amount of roadside land use expressed as percentage of land use adjacent to
the road, taken from ARMS Digital Image data. An estimate of landuse was
recorded in 4-band intervals every 1.0 km intervals along the entire length of the
link. An average value for the link was then recorded for each strategic link. The
LU values were:
Land Use
Intensity
Low
Medium
High
Urban Area

Land Use
Code
1
2
3
4

Estimated
Land Use
0-29%
30-59%
60-89%
>90%

LU
Factor
0.15
0.45
0.75
0.95

E3.2 Free Flow Speeds for Remainder of the Network

E3.2.1 Toll Roads and Toll Road Access Link Free Flow Speeds
A maximum speed limit of 100 km/h applies n Indonesia. A free flow speed of 90 km/h for all
vehicles was adopted. For toll road access links, the same speed as urban area links was used.
E3.2.2 Free Flow Speeds for Other Links

Carl Bro International a|s and Associates

E- 8

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

A free flow speed of 50 km/h was assumed for other links in the network validation model, and
model results verified this assumption. Therefore, for all urban area links and corridors a free
flow speed of 50 km/h was adopted. The same speed was also adopted for rural area non-strategic
links.

E4. SPEED FLOW CURVES

The JARNS speed flow curves were defined to have three points (see Figure E.1), based on the
Volume Capacity Ratio (VCR) and the Free Flow Speed (FFS):
1. VCR = 0.0, Speed (V0) = FFS;
2. VCR=0.85, Speeds (V85) as per IHCM relationships by road type, and
3. Speed (V10)= 10 km/h, VCR back calculated from the IHCM relationships.
Figure E.1 General Form of the Speed Flow Curve
Speed
V0=FFS

10 km/h

VCR = 0.85

VCR

The derivation of points 1 and 2 is straight forward as shown below. The third point pre-defines
the speed for which the VCR needs to be estimated. This speed is usually referred to as minimum
or cut-off speed. i.e. no link in the network will be allowed to have a speed less than this value,
irrespective of the traffic volume or VCR. Initially a speed of 6kph was used in the model
validation process, but it was realised that it is too low a cut-off speed for inter-urban roads and
would not be sustained, as the peak spreading would happen and higher minimum speed could be
attained. A value of 10 km/h was used, and the base network validation gave better estimates of
base year journey times than those with 6 km/h minimum speed. Therefore, minimum speed of 10
km/h was adopted.
1.

Speed (V0) at VCR=0,

is the same as FFS, i.e. Vo=FFS

2.

Speeds (V85) at VCR = 0.85

Speed (V85) for VCR = 0.85 uses the same relationship for all road types and for FFS <100.
V85 = FFS- 0.85 * [CO * (C1 - C2 * (100 - FFS) - C3 * ((100-FFS)^C4))]

(1)

The FFS for each link is derived as described above, and the values of parameters C1, C2, C3 and
C4 are given in the table below.

Carl Bro International a|s and Associates

E- 9

Java Arterial Road Network Study (JARNS)

3.

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

Speeds at VCR > 0.85

Speed for VCR > 0.85 are computed from according to the road type:
For Road Type 2: 2-Lane Single Carriageway 2-way:
VCRx = V85 - [VCR - 0.85] * [C5 - C6*(56 - V85) + (C7 * (56 - V85)^C8)]

(2)

For Road Types 3, 4 and 5: 4+Lanes Single or Divided Carriageways, 2-way:

VCRx = V85 - [VCR - 0.85] * [C5 - C6*(72.6 - V85) + (C7 * (72.6 - V85)^C8)] (3)
To determine the VCR at a speed of 10 km/h (VCR10) the above equations 2 and 3 can be rewritten as equations 4 and 5 below:
VCR10 = 0.85 + {(V85 - 10) / [C5 - C6 * (56- V85) + (C7*((56- V85)^C8)]}

(4)

VCR10 = 0.85 + {(V85 - 10) / [C5 - C6 * (72.6- V85) + (C7*((72.6- V85)^C8)]}

(5)

Where C1..C8 parameters are given in table below:

Road Type
2
3 / 4 /5

CO
1.0
0.9

C1
51.76
35.88

C2
0.505
0.400

C3
0.000156
0.000040

C4
2.5
1.2

C5
80.03
90.35

C6
1.75
1.75

C7
0.0026
0.0026

C8
2.42
2.42

Applying equations 1, 4, and 5 and using the FFS in bands of 5 km/h 19 speed flow curves were
defined, 10 curves for road type 2 and 9 curves for road types 3/4/5 combined, respectively.
These curves are tabulated below in Table E.5 and E.6 and shown in Figures E.2 and E.3 for road
type 2 and 3/4/5 respectively.
Table E.5: Speed Flow Curves for Road Type 2 (2-Lane Single Carriageway)
Curve
Number

Capacity Index
Code
(Network Model)

Free Flow Speed

(FFS) km/h

Speed at VCR=0.85
(V85)

VCR at 10 km/h
(VCR10)

1
2
3
4
5
6
7
8
9
10

1
2
3
4
5
6
7
8
9
10

30
35
40
45
50
55
60
65
70
75

21
23
25
28
30
32
35
37
40
42

1.18
1.23
1.26
1.31
1.33
1.35
1.38
1.39
1.40
1.41

Carl Bro International a|s and Associates

E- 10

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

Figure E.2: Speed Flow Curves for Road Type 2 (2-Lane Single Carriageway)
80
70

Speed (km /h)

60
50
40
30
20
10
0
0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

Volum e Capacity Ratio (VCR)

Table E.6: Speed Flow Curves for Road Types 3/4/5 (Multi-lane Carriageways)
Curve
Number

Capacity Index
Code
(Network Model)

Free Flow Speed

(FFS) km/h

Speed at VCR=0.85
(V85)

VCR at 10 km/h
(VCR10)

11
12
13
14
15
16
17
18
19

20
21
22
23
24
25
26
27
28

50
55
60
65
70
75
80
85
90

38
41
45
48
52
55
59
62
66

1.49
1.52
1.55
1.56
1.57
1.57
1.57
1.57
1.56

Carl Bro International a|s and Associates

E- 11

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

Figure E.3: Speed Flow Curves for Road Types 3/4/5 (Multi-lane Carriageways)
90
80

Speed (km /h)

70
60
50
40
30
20
10
0
0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

Volum e Capacity Ratio (VCR)

E4.1 Application of Speed Flow Relationships

E4.1.1 Strategic Network Links
Capacity and free flow speed were calculated for all links in the strategic network, as a function
of physical, geometric, land use and other characteristics, using the previously defined
relationships. A specific speed flow curve number was allocated based on road type, and
estimated FFS rounded to the nearest 5 km/h. This link code is used by the network assignment
model through the Capacity Index Codes as given in tables E.5 and E.6 above, to calculate the
speed for a given volume of traffic.
For the forecast year scenario, if the link is upgraded, link capacity and FFS are recalculated. A
new Capacity Index code is allocated and coded into the network model according to the new
FFS and the Road Type.
E4.1.2 Non-Strategic Links
Toll Road Speed Flow Curve
A single speed flow curve was defined for all toll road links with a FFS of 90km/h. The curve has
the same characteristics as a multi-lane carriageway curve number 9, described above in Table
E.6. Capacity index code number 30 was allocated to all toll road links. (See summary Table E.7)
Toll Road Access Links Speed Flow Curve
A single speed flow curve was defined for all toll road access links with a FFS of 50km/h. The
curve has the same characteristics as the multi-lane carriageway curve number 1, and described
above in Table E.6. Capacity index code number 29 was allocated to all toll road access links.
(See summary Table E.7)

Carl Bro International a|s and Associates

E- 12

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix E:

Road Capacity and Speed Flow Relationships

Urban Area Links and Corridors Speed Flow Curves

A single speed flow curve was defined for all urban area links and corridors with a FFS of
50km/h. The curve has the same characteristics as the 2-lane single carriageway curve number 5,
and described above in Table E.5. Capacity index code numbers 11, 12, and 13 were allocated to
these urban road links and corridors depending upon their road type allocation. (See summary
Table E.7)
Rural Roads Speed Flow Curves
A single speed flow curve was defined for all these low capacity roads with a FFS of 50km/h. The
curve has the same characteristics as the 2-lane single carriageway curve number 5, and described
above in Table E.5. Capacity index code number 14 was allocated to all these links. (See
summary Table E.7)
Table E.7: Speed Flow Curves for Non-Strategic Roads
Road
Type
Number

6/ 7/ 8
9
11
12
13
14

Road Type
Description

Speed
Flow
Curve
Number

Capacity
Index
Code
(Network
Model)

Free
Flow
Speed
(FFS)
km/h

Speed at
VCR=0.85
(V85)

VCR at
10 km/h
(VCR10)

Toll Roads
Toll Road Access Links
IRMS Urban Links
IRMS Kota Links
Urban Corridors
Rural Roads

19
11
5
5
5
5

29
30
11
12
13
14

90
50
50
50
50
50

66
38
30
30
30
30

1.56
1.49
1.33
1.33
1.33
1.33

Carl Bro International a|s and Associates

E- 13

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

APPENDIX F:
ASSIGNMENT VALIDATION

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix F:

Assignment Validation

F.1 INTRODUCTION
The JARNS base year (2000) network assignment validation is discussed in Section 5 of this
report. This appendix elaborates the assignment validation results. The assignment results are
presented for each link of the 29 cordons and screenlines defined for the network validation. The
cordons and screenlines are shown in Figure 5.9 in Section 5, and the results are presented in
Table F.1 below.
Table F.1: Base Year (2000) Assignment Validation Results by Each Link
Cordon or Screenline Description
No.

Anode

Bnode

Link Description

Count
Source

20230

20222

Bayah - Malingping

Estimate

110

106

-4

-4%

20211

20213

Cipanas - Rangkasbitung

JARNS

91

116

25

27%

22030

21932

Sp. Bitung - Cikande

JARNS

913

878

-35

-4%

28130

28122

Toll Balaraja Brt - Ciujung

Jasa Marga

1,723

1,660

-63

-4%

2,837

2,760

-77

-3%

Screenline Total

Obs.

Assign

A-O

A/O
(%)

27204

20512

Sukabumi - Cianjur

631

546

20370

20390

Ciawi - Puncak (Cianjur)

JARNS

1,055

1,080

25

2%

20516

20520

Cileungsi - Cianjur

Estimate

481

460

-21

-4%

28070

28080

Toll Cikarang - Karawang Tmr

Jasa Marga

3,954

4,252

298

8%

21820

21712

Cikarang - Karawang

ATC

702

597

6,823

6,935

112

2%

JARNS

1,193

1,231

38

3%
9%

JARNS

Screenline Total

-85 -13%

-105 -15%

20520

20631

21721

21611

Cikampek - Sadang

JARNS

1,706

1,868

162

21720

21520

Cikampek - Pamanukan

JARNS

1,415

1,516

101

7%

4,314

4,615

301

7%

20

2%

Selajambe - Padalarang

Screenline Total

20652

20720

Cilaunyi - Nagreg

JARNS

808

828

21312

21314

Sumedang - Wado

JARNS

173

147

-26 -15%

21311

21320

Sumedang - Cijelag

JARNS

608

735

127

21%

21520

21430

Pamanukan - Janggaa

JARNS

1,379

1,406

27

2%

2,968

3,116

148

5%

800

798

-2

0%

Screenline Total

20914

20913

Tasikmalaya - Ciamis

ATC

21311

21320

Sumedang - Cijelag

JARNS

608

735

127

21%

21520

21430

Pamanukan - Janggaa

JARNS

1,379

1,406

27

2%

2,787

2,939

152

5%

Screenline Total

30234

30230

Kr. Pucung - Wangon

JARNS

402

496

94

23%

21130

32921

Losari - Pejagan

JARNS

1,581

1,612

31

2%

1,983

2,108

125

6%

JARNS

687

772

85

12%

Screenline Total

30223

30531

30430

30412

Kelampok - Banjanegara

ATC

453

351

32730

32731

Bantarbolang - Randudongkal

JARNS

332

384

Buntu - Gombong

Carl Bro International a|s and Associates

-102 -23%
52

16%

F- 1

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix F:

Assignment Validation

Table F.1: Base Year (2000) Assignment Validation Results by Each Link
Cordon or Screenline Description
No.

Anode

Bnode

Link Description

Count
Source

32713

32714

Tegal - Pemalang

JARNS

Obs.

Assign

A/O
(%)

A-O

1,441

1,400

-41

-3%

2,913

Screenline Total

2,907

-6

0%

76

11%

30511

30520

Kebumen - Prembun

JARNS

682

758

30411

30731

Banjanegara - Kaliwiro

Estimate

405

330

32530

32531

Blado - Bawang

JARNS

85

86

1%

32510

32511

Batang - Banyuputih

JARNS

1,290

1,543

253

20%

2,462

2,717

255

10%

998

806

Screenline Total

-75 -19%

10

40201

40315

Yogya - Wonosari

ATC

10

31012

31010

Prambanan - Klaten

JARNS

1,207

1,316

109

9%

10

30920

30921

Boyolali - Kartosuro

JARNS

981

1,160

179

18%

10

31541

31540

Gubug - Godong

JARNS

655

642

-13

-2%

10

32140

32113

Semarang - Demak

JARNS

1,523

1,480

-43

-3%

5,364

5,404

40

1%

10

Screenline Total

11

31232

31230

11

31211

11

-192 -19%

Baran - Pracimantoro

Estimate

130

168

38

29%

31220

Sukoharjo - Wonogiri

Estimate

405

410

1%

31311

31320

Karanganyar - Bts. Jatim

JARNS

328

402

74

23%

11

31410

31411

Sragen - Widodaren

Estimate

568

610

42

7%

11

31511

31520

Purwodadi - Wirosari

Estimate

259

574

315 122%

11

31812

31830

Juwana - Rembang

Estimate

730

866

136

19%

2,420

3,030

610

25%

11

Screenline Total

12

50430

50412

Trenggalek - Tulungagung

JARNS

464

508

44

9%

12

51912

51911

Caruban - Nganjuk

JARNS

611

560

-51

-8%

12

52220

52433

Bojonegoro - Babat

JARNS

352

680

328

93%

12

52330

52313

Glandong - Tuban

JARNS

829

764

-65

-8%

2,256

2,512

256

11%

13

50730

50713

Blitar - Kepanjen

JARNS

590

608

18

3%

13

50743

50742

Kandangan - Batu

JARNS

300

282

-18

-6%

13

51715

51720

Jombang - Mj. Agung

JARNS

1,404

1,418

14

1%

13

52411

52413

Lamongan - Gresik

JARNS

966

1,120

154

16%

13

52421

52520

Sadang - Gresik

JARNS

435

424

-11

-3%

3,695

3,852

157

4%

590

608

18

3%

12

Screenline Total

13

Screenline Total

14

50730

50713

Blitar - Kepanjen

JARNS

14

50742

50740

Batu - Malang

Estimate

300

282

-18

-6%

14

58160

51411

Gempol - Pandaan

JARNS

2,162

2,124

-38

-2%

14

58150

51413

Gempol - Pasuruan

JARNS

2,061

2,008

-53

-3%

5,113

5,022

-91

-2%

14

Screenline Total

15

50830

50812

Turen - [Link]

JARNS

359

296

15

51330

51331

Pasuruan - Pilang

JARNS

1,567

1,560

Carl Bro International a|s and Associates

-63 -18%
-7

0%

F- 2

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix F:

Assignment Validation

Table F.1: Base Year (2000) Assignment Validation Results by Each Link
Cordon or Screenline Description
No.

Obs.

Assign

A/O
(%)

A-O

Anode

Bnode

1,926

1,856

-70

-4%

16

50820

50822

Wonorejo - Pondok Dalem

JARNS

371

459

88

24%

16

57404

51320

Probolinggo - Baduan

JARNS

747

792

45

6%

1,118

1,251

133

12%

1,723

1,660

-63

-4%

15

Link Description

Count
Source

Screenline Total

16

Screenline Total

17

28130

28122

Toll Balaraja Brt - Ciujung

Jasa Marga

17

22030

21932

Sp. Bitung - Cikande

JARNS

913

878

-35

-4%

17

20211

20213

Cipanas - Rangkasbitung

JARNS

91

116

25

27%

17

20402

20370

Ciawi - Cibadak

JARNS

1,069

1,084

15

1%

17

20390

20370

Ciawi - Puncak (Cianjur)

JARNS

1,055

1,080

25

2%

17

20516

20520

Cileungsi - Cianjur

Estimate

481

460

-21

-4%

17

28080

28070

Toll Cikarang - Karawang Tmr

Jasa Marga

3,954

4,252

298

8%

17

21712

21820

Cikarang - Karawang

ATC

702

597

9,988

10,127

139

1%

JARNS

1,193

1,231

38

3%

17

Cordon Total

-105 -15%

18

20520

20631

18

20720

20652

Cilaunyi - Nagreg

JARNS

808

828

20

2%

18

21313

21330

Cileunyi - Sumedang

JARNS

631

660

29

5%

18

21530

20640

Cagak - Bandung

JARNS

280

284

1%

18

20633

20632

Cisomang - Padalarang

JARNS

961

1,040

79

8%

3,873

4,043

170

4%

465

50%

18

Selajambe - Padalarang

Cordon Total

19

21420

27406

Karangampel - Cirebon

JARNS

921

1,386

19

21120

27404

Palimanan - Cirebon

JARNS

1,727

1,186

-541 -31%

19

28220

28210

Ciperna - Plumbon

Jasa Marga

717

1,580

863 120%

19

21011

21020

Cirebon - Kuningan

JARNS

856

754

-102 -12%

19

21130

27402

Cirebon - Losari

JARNS

1,204

1,558

20

32511

32510

Batang - Banyuputih

JARNS

20

32530

32512

Bandar - Batang

JARNS

20

32611

37501

Wiradesa - Pekalongan

Pek / Prob

19

20

6,464 1,039

19%

1,290

1,543

253

20%

85

98

13

15%

1,652

1,770

118

7%

3,027

Cordon Total

29%

5,425

Cordon Total

354

3,411

384

13%

-137

-7%

21

32414

32413

Kendal - Semarang

JARNS

1,881

1,744

21

32425

32213

Cangkiran - Ungaran

JARNS

192

142

21

32210

32213

Ungaran - Bawen

JARNS

2,032

2,306

21

32131

32130

Semarang - Godong

JARNS

655

574

21

32113

32140

Semarang - Demak

JARNS

1,523

1,480

-43

-3%

6,283

6,246

-37

-1%

998

806

21

Cordon Total

-50 -26%
274

13%

-81 -12%

22

40315

40201

Yogya - Wonosari

ATC

22

31010

31012

Prambanan - Klaten

JARNS

1,207

1,316

109

9%

22

30813

30851

Magelang - Muntilan

JARNS

1,209

1,220

11

1%

Carl Bro International a|s and Associates

-192 -19%

F- 3

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix F:

Assignment Validation

Table F.1: Base Year (2000) Assignment Validation Results by Each Link
Cordon or Screenline Description
No.

Anode

Bnode

Link Description

Count
Source

22

30621

40103

Purwodadi - Wates

ATC

22

Obs.

Assign

A/O
(%)

A-O

601

676

75

12%

4,015

Cordon Total

4,018

0%

23

31320

31311

Karanganyar - Bts. Jatim

JARNS

328

402

74

23%

23

31411

31410

Sragen - Widodaren

Estimate

568

610

42

7%

23

31530

31531

Purwodadi - Gundih

JARNS

261

252

-9

-3%

23

30920

30921

Boyolali - Kartosuro

JARNS

981

1,160

179

18%

23

31020

30923

Klaten - Kartosuro

ATC

1,109

994

23

31121

37202

Surakarta - Sukoharjo

JARNS

887

856

-31

4,134

4,274

140

3%

24

52421

52520

Sadang - Gresik

JARNS

435

424

-11

-3%

24

52411

52413

Lamongan - Gresik

JARNS

966

1,120

154

16%

24

51715

51720

Jombang - Mj. Agung

JARNS

1,404

1,418

14

1%

24

51411

58160

Gempol - Pandaan

JARNS

2,162

2,124

-38

-2%

24

51413

58150

Gempol - Pasuruan

JARNS

2,061

2,008

-53

-3%

24

52630

57801

Madura Ferry

Operator

296

336

40

14%

23

Cordon Total

24

Cordon Total

-115 -10%
-3%

7,324

7,430

106

1%

JARNS

25

58150

51413

Gempol - Pasuruan

2,061

2,008

-53

-3%

25

58160

51411

Gempol - Pandaan

JARNS

2,162

2,124

-38

-2%

25

50742

50740

Batu - Malang

Estimate

300

282

-18

-6%

25

50712

57303

Malang - Kepanjen

JARNS

584

580

-4

-1%

25

50720

57300

Malang - Turen

JARNS

516

510

-6

-1%

25

51331

51330

Pasuruan - Pilang

JARNS

1,567

1,560

-7

0%

7,190

7,064

-126

-2%

45

6%

25

Cordon Total

26

51320

57404

Probolinggo - Baduan

JARNS

747

792

26

51330

51331

Pasuruan - Pilang

JARNS

1,567

1,560

-7

0%

26

50821

51311

Probolinggo - Grobogan

JARNS

789

740

-49

-6%

3,103

3,092

-11

0%

26

Cordon Total

27

7111

27703

Sumatra Ferry

Operator

466

486

20

4%

28

52630

57801

Madura Ferry

Operator

296

336

40

14%

29

6111

51013

Bali Ferry

Operator

262

248

-14

-5%

Carl Bro International a|s and Associates

F- 4

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

APPENDIX G:
TRIP END MODELS

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix G:

Trip End Models

G1. INTRODUCTION
G1.1 Background
The trip end model development methodology is based on the zonal regression analyses, and has
been described in Section 6.2. This Appendix provides further details on the model estimation
process, and lists the models used for the estimation of forecast year trip ends. A number of
different types and forms of relationships were tried and rejected. The general form of the model
finally found to give the best results was:
TEi = a * V1b * V2c * V3d
where: TEi : is the Number of Trip Ends in Zone i
V1, V2 & V3 are independent variables, and
a, b, c, & d are parameters to be calibrated.
As discussed in Section 6.2, trip end models were calibrated for the whole of Java, but the results
were not acceptable, due to poor correlation. Therefore, three separate models were developed for
Jakarta and West Java; Central Java and Yogyakarta, and East Java.
G1.2 Independent Variables
A wide range of zonal data was collected in JARNS, as reported in Technical Report No.9
Economic and Regional Development in Java. The data was prepared for each of the 262 study
area zones for the variable listed below, under four categories:

Zonal Population
o
Total Population
o
Urban Population
o
Rural Population
o
Density (Population / sq-km)
Zonal GDP
o
Total GDP, and GDP by the following sub-sectors:
o
Agriculture
o
Industry (manufacturing)
o
Services
o
Mining
o
GDP /Capita
Zonal Employment
o
Total Employment and by the following classifications:
o
Agriculture
o
Industry (manufacturing)
o
Services
o
Mining
Vehicle Registration Data

Analysis of these variables showed that in many cases the data is not very reliable or have been
derived from other variables listed above. This can be seen from the high correlation between a
number of independent variables, presented in Table G.1 below.

Carl Bro International a|s and Associates

G- 1

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix G:

Trip End Models

Table G.1: Independent Variable Correlation Matrix

Variable

Total
Pop

Urban
Pop

Rural
Pop

Pop
Dens

Total
GDP

Agri.
GDP

Man.
GDP

Serv.
GDP

Min.
GDP

GDP/
Cap

Tot
Emp.

Agric.
Emp.

Man
Emp.

Serv
Emp.

Min.
Emp.

Total Pop
Urban Pop
Rural Pop
Pop Dens
Total GDP
Agri. GDP
Man. GDP
Serv. GDP
Min. GDP
GDP/Capita
Tot Emp.
Agric. Emp.
Man Emp.
Serv Emp.
Min. Emp.

1.00
.81
.48
.44
.71
.47
.63
.64
.30
.25
.96
.46
.86
.89
.39

1.00
-.13
.74
.85
-.09
.79
.82
.11
.43
.71
-.04
.83
.82
.50

1.00
-.37
-.08
.93
-.12
-.14
.35
-.23
.55
.83
.21
.27
-.09

1.00
.70
-.31
.59
.73
-.02
.54
.35
-.27
.47
.53
.37

1.00
-.03
.89
.97
.12
.69
.65
-.06
.67
.83
.23

1.00
-.11
-.09
.29
-.18
.56
.82
.18
.29
-.07

1.00
.77
.09
.78
.55
-.12
.70
.69
.32

1.00
.09
.61
.59
-.10
.58
.80
.18

1.00
.06
.30
.25
.24
.23
.03

1.00
.20
-.23
.30
.37
.09

1.00
.61
.82
.86
.40

1.00
.23
.19
.23

1.00
.75
.51

1.00
.12

1.00

The variables with a high correlation with the trip ends were Total Zonal GDP, Total Population
and Urban Population. Using these variables trip end models were calibrated for each of the five
vehicle categories, using step-wise multiple-linear regression technique available in the SPSS
software package.

G2. CAR TRIP END MODELS

The stepwise regression analysis included 3 independent variables for the car trip end models.
These were: Total Population (TPOP), Urban Population (UPOP) and Total GDP (TGDP). The
models showed that the coefficient for the total population was negative, which is counterintuitive. Therefore, in the final models total population was not included as the independent
variable. The final models for each province are given below, and each variable represents the
zonal values. The model for the Car Trip Ends took the form:
b

CAR_TEi = a * TGDP * UPOP

where: CAR_TEi : is the Number of Car Trip Ends in Zone-I,

TGDP is the Total Zonal GDP in Zone I;
UPOP is the Urban Population in Zone I, and
a, b, & c are the estimated model parameters, and (t statistics) are given in the
table below:
Model Region
1. West Java and Jakarta
2. Central Java and Yogyakarta
3. East Java

Parameter Value /
t-statistic
Parameter Value
t-statistic
Parameter Value
t-statistic
Parameter Value
t-statistic

Carl Bro International a|s and Associates

Parameter Values/ Statistics

a
b
c
0.116
0.620
0.0418
-1.683
4.734
0.478
0.0037
0.913
0.026
-3.189
5.228
0.266
0.015
0.663
0.199
-3.242
4.850
2.498

R2
0.41
0.33
0.54

G- 2

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix G:

Trip End Models

G3. LGV TRIP END MODELS

The stepwise regression analysis for the LGV trip end calibration included 3 independent
variables in the models. These were: Total Population (TPOP), Urban Population (UPOP) and
Total GDP (TGDP). The model showed that the coefficient for the total population is negative,
which is counter-intuitive, and the coefficient for the Urban population was not significant.
Therefore, in the final models total population and the urban population were not included as the
independent variables. The final models are therefore function of zonal GDP alone, and are given
below:
LGV_TEi = a * TGDP

where: LGV_TEi : is the Number of LGV Trip Ends in Zone-I,

TGDP is the Total Zonal GDP in Zone I; and
a, & b are the estimated model parameters, and (t statistics) are given in the table
below:
Parameter Value /
t-statistic

Model Region

Parameter Value
t-statistic
Parameter Value
t-statistic
Parameter Value
t-statistic

1. West Java and Jakarta

2. Central Java and Yogyakarta
3. East Java

Parameter Values/ Statistics

a
b
0.088
0.609
-1.959
6.676
0.0141
0.767
-3.336
7.598
0.189
0.561
-1.701
7.377

R2
0.37
0.37
0.39

G4. MGV TRIP END MODELS

The MGV models include only one independent variable, Total zonal GDP (TGDP). No other
variable showed significant correlation to the zonal trip ends. The MGV models are:
MGV_TEi = a * TGDP

where: MGV_TEi : is the Number of MGV Trip Ends in Zone-I,

TGDP is the Total Zonal GDP in Zone I; and
a, & b are the estimated parameters, and (t statistics) are given in the table below:
Model Region
1. West Java and Jakarta
2. Central Java and Yogyakarta
3. East Java

Parameter Value /
t-statistic
Parameter Value
t-statistic
Parameter Value
t-statistic
Parameter Value
t-statistic

Carl Bro International a|s and Associates

Parameter Values/ Statistics

a
b
0.1021
0.632
-2.138
8.060
0.00205
0.928
-4.491
8.523
0.0429
0.688
-2.834
7.979

R2
0.46
0.43
0.43

G- 3

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix G:

Trip End Models

G5. HGV TRIP END MODELS

Similar to the MGV models, the HGV models include only one independent variable, Total zonal
GDP (TGDP). No other variable showed significant correlation to the zonal trip ends. The HGV
models are:
HGV_TEi = a * TGDP

where: HGV_TEi : is the Number of HGV Trip Ends in Zone-I,

TGDP is the Total Zonal GDP in Zone I; and
a, & b are the estimated model parameters, and (t statistics) are given in the table
below:
Parameter Value /
t-statistic

Model Region
1. West Java and Jakarta
2. Central Java and Yogyakarta
3. East Java

Parameter Value
t-statistic
Parameter Value
t-statistic
Parameter Value
t-statistic

Parameter Values/ Statistics

a
b
0.00295
0.775
-3.653
6.615
0.0137
0.626
-2.956
5.465
0.00292
0.801
-4.444
7.854

R2
0.37
0.24
0.42

G6. BUS TRIP END MODELS

The regression models for the Bus (vehicle trips) were found to be similar to those calibrated for
the Car trip ends. Total population as an independent variable with negative coefficient was not
included in the models. The two significant independent variables were: Urban population and
Total zonal GDP. The models were:
b

BUS_TEi = a * TGDP * UPOP

where: BUS_TEi : is the Number of BUS Trip Ends in Zone-I,

TGDP is the Total Zonal GDP in Zone I;
UPOP is the Urban Population in Zone I, and
a, b, & c are the estimated parameters, and (t statistics) are given in the table
below:
Model Region
1. West Java and Jakarta
2. Central Java and Yogyakarta
3. East Java

Parameter Value /
t-statistic
Parameter Value
t-statistic
Parameter Value
t-statistic
Parameter Value
t-statistic

Carl Bro International a|s and Associates

Parameter Values/ Statistics

a
b
c
0.00026
0.858
0.0575
-3.862
3.935
0.394
0.0009
0.864
0.047
-3.721
4.611
0.448
0.129
0.273
0.239
-1.229
1.550
2.327

R2
0.32
0.29
0.24

G- 4

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix G:

Trip End Models

G7. FORECAST YEAR TRIP ENDS

The method used to estimate the future year zonal trip ends then took the form:
TEFY = TEBY * (S_TEFY / S_TEBY)
where: TEFY are Future Year Trip Ends for Zone i
TEBY are the Base Year Trip Ends in Zone i
S_TEFY are the Synthesised Future Year Trip Ends for Zone i
S_TEBY are the Synthesised Base Year Trip Ends for Zone I
(both the S_BY and S_TEFY were estimated using the regression models detailed above)
This process yielded the zonal trip ends that reflected the effects of estimated changes in socioeconomic data between the base and the forecast years. This overall growth in traffic was
estimated exogenously, and has been described in Appendix F, Traffic Growth Factors. The
future year trip ends (TEFY) estimated from the above models were then controlled to the overall
traffic growth volumes estimated using the traffic growth factors. Similarly, the trip ends for the
two external zones (Bali and Sumatra) were directly computed using the growth in regional GDP.

Carl Bro International a|s and Associates

G- 5

Java Arterial Road Network Study (JARNS)

Technical ReportNo.10:
Network Model Development and Application

APPENDIX H:
TRAFFIC GROWTH FACTORS

Carl Bro International a|s and Associates

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix H:

Traffic Growth Factors

H1. INTRODUCTION
Two approaches have been adopted to forecast the growth of traffic on the road network. The
detailed spatial analyses set out in Appendix G have been used to provide disaggregated
forecasts of the demand for transport at a zonal level, to represent effects of changes in spatial
data. However to provide an overall picture of the likely growth in road traffic in Java and to
provide control totals for the zonal forecasts, estimates have been made of growth both on a
Java-wide basis and at a provincial level. This Appendix presents this latter analysis.
H1.1 Scope of the Forecasts
The different components of the traffic streams will have different growth rates. Forecasts
have therefore been made for following categories of vehicles:

Motorcycles (Vehicle Type 1)

Cars (Vehicle Type 2)
Light Goods Vehicles LGV (Vehicle Type 4)
Angkots (Vehicle Type 3)
Buses (Vehicle Types 5 & 6)
Trucks (MGV & HGV - Vehicle Types 7,8,9,10,11)
Total Traffic Combined (Vehicle Types 1-to-11)

Forecasts for each of these categories of vehicles have been made for the following regions:

Jakarta, Banten, and West Java

Central Java and Yogyakarta

East Java

Java as a whole
In addition, specific forecasts of growths have been made for vehicular ferry traffic:

Java Sumatra (Merak Bakauheni), and

Java Bali (Banyuwangi-Gilmanuk)

H1.2 Forecast Scenarios

Two sets of traffic forecasts have been produced. The first set relates to the central growth
scenario of GRDP growth forecasts, and the resultant traffic growth forecasts are presented in
Section H3. The second set represents an optimistic GRDP growth, and the related traffic
growth forecasts are presented in Section H4.

Carl Bro International a|s and Associates

H- 1

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix H:

Traffic Growth Factors

H2. KEY DETERMINANTS OF TRAFFIC GROWTH

H2.1 Introduction
The key determinants of traffic growth are population growth and GRDP growth.
forecasts on which the analysis has been based are set out in Table H.1.

The

Table H.1: Forecasts of GRDP and Population Growth for Java 2000-2020
GRDP at constant 1993 prices (Rp Trillions)
Region
2000
2010
2020
(Provinces)
Total
Index
Total
Index
Total
Index
East Java
Central Java
Yogyakarta
Central Java and Yogyakarta
West Java
Jakarta
West Java and Jakarta
Java
Sumatra
Bali

56.6
37.5
4.8
42.3
69.6
60.9
130.5
229.4
85.3
7.4

1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00

86.0
55.2
7.4
62.6
119.2
101.3
220.5
369.2
142.9
12.5

East Java
Central Java
Yogyakarta
Central Java and Yogyakarta
West Java
Jakarta
West Java and Jakarta

34.5
30.9
3.1
34.0
43.6
8.4
51.9

1.00
1.00
1.00
1.00
1.00
1.00
1.00

36.3
32.3
3.3
35.5
47.7
9.8
57.6

120.4

1.00

129.4

1.52
1.47
1.54
1.48
1.71
1.66
1.69
1.61
1.68
1.69

142.7
88.5
11.9
100.4
234.2
166.9
401.0
644.1
238.5
20.9

2.52
2.36
2.48
2.37
3.37
2.74
3.07
2.81
2.79
2.84

1.052
1.046
1.048
1.046
1.095
1.172
1.110

38.0
33.4
3.3
36.8
50.0
10.7
60.7

1.101
1.084
1.075
1.083
1.147
1.277
1.170

1.075

135.4

1.125

POPULATION (millions)

Java

H2.2 Forecasts of Road Traffic Growth for Java as a Whole

H2.2.1 Total Traffic
Forecasts of traffic growth for Java as a whole have been made based on the forecast growth
in GRDP and a multiplier of 1.15. Given the limited confidence that can be attached to much
of the traffic data that is available, it is difficult to derive a precise estimate of the multiplier.
However, previous studies have indicated that the elasticity of vehicle use to GRDP appears
to lie within the range of about 1.1 (CAPEX 2) to 1.5 or even 2.0, although there has been a
higher incidence of estimates at the lower end of the range than the upper-end.
For the purpose of JARNS, it has been assumed that a factor of 1.15 would be appropriate.
This is somewhat below the figure of 1.25 used in IRMS98 and the Screening Study, but is
broadly consistent with recent (pre-crisis) experience in the instances where reasonably
reliable traffic figures exist. The comparison between the traffic flows on the Citarum Bridge
and GRDP to which the multiplier has been applied is set out in Table H.2. This indicates
that based on, admittedly limited evidence the use of a value of 1.15 appears to give
reasonable, although possibly conservative results.

Carl Bro International a|s and Associates

H- 2

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix H:

Traffic Growth Factors

Table H.2: Citarum Bridge and Mojokerto Bypass Observed Traffic Growth and Local
GRDP Growth (Per cent per year)
Year

Observed
Traffic Growth

1993
1994
1995
1996
1997

Provincial
GRDP Growth

2.1%
11.7%
10.2%
10.1%
8.0%

Provincial GRDP Growth

Increased by 15 per cent

6.9%
8.2%
8.8%
8.3%
5.1%

7.9%
9.4%
10.1%
9.6%
5.8%

The use of this multiplier and the GRDP (central scenario) growth factors give the following
estimates for the total traffic growth in Java to 2010 and 2020 as follows:

Total Traffic Growth factor 2000 to 2010

Total Traffic Growth factor 2000 to 2020

1.72
3.26

H3. GRDP CENTRAL GROWTH SCENARIO TRAFFIC FORECASTS

H3.1 Java Wide Growth Factors by Vehicle Types
In addition to the forecasts of total traffic, estimates for Java as a whole have been made of
the individual components of the total traffic flows by following categories of vehicles:

Motorcycles
Cars
Angkots
Light Goods Vehicles (LGV)
Buses
Trucks (MGV & HGV)

Information is not available directly on the growth of the traffic flows for each of these
vehicle types, but figures on vehicle ownership, that are available have been used as a proxy
for this. Forecasts for future levels of vehicle ownership have been based on a series of
simple linear regressions for the individual vehicle types linking growth in vehicle ownership
and implicitly in traffic growth to growth in GRDP or population over the period 1990-1997.
The GRDP or population multipliers derived are summarised below in Table H.3.
Table H.3: Relationship between GRDP or Population Growth and Changes in Vehicle
Ownership used in the Derivation of Traffic Forecasts
Vehicle Category
Cars
Motorcycles
Angkots (Based on Population)
LGV, MGV & HGV
Buses

Growth as Proportion of GRDP

1.85
1.85
1.00
1.11
1.00

Carl Bro International a|s and Associates

R2
0.975
0.995
N/A
0.993
0.887

H- 3

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix H:

Traffic Growth Factors

Applying these regressions coefficients and controlling to the overall traffic growth for Java
gives the following growth factors and vehicle proportion splits for 2010 and 2020 for Java as
a whole. The growth factors for key vehicle categories for the whole of Java are shown below
in Figure H.1 and detailed in Table H.4.
Figure H.1: Central Growth Scenario Road Traffic Growth Forecasts - Java

Central Scenario Traffic Growth Forecast

5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
2000

2005
Cent-GRDP

2010
Cars

2015
Trucks

2020

Buses

Table H.4: Central Growth Scenario Road Traffic Growth Forecasts - Java
Vehicle Category
Motorcycles
Cars
LGV
Angkots
Buses
Trucks (MGV&HGV)
Total (All Vehs)
GRDP Growth

Traffic
Proportions
In 2000
35%
27%
9%
11%
4%
14%
100%

Growth to
2010
1.93
1.93
1.52
1.07
1.46
1.52
1.72

Traffic
Proportions
in 2010
40%
30%
8%
7%
3%
12%
100%
1.61

Growth to
2020
3.91
3.91
2.70
1.12
2.53
2.70
3.26

Traffic
Proportions
in 2020
42%
32%
7%
4%
3%
12%
100%
2.81

Within the total traffic, the shares of private vehicles, car and motorcycles are forecast to
increase. With increasing affluence, the demand for transport will grow, but it is expected that
an increasing share of this demand will be met by private transport, as ownership of private
cars and motorcycles increases. Although the total flows of Angkots and buses are expected
to grow, the rate of increase is below the average for the traffic as a whole, and their share of
the total consequently falls. Again, although the volume of truck traffic is forecast to increase
significantly, the rate of growth will again be below the average for the traffic as a whole, and
the share of truck traffic will decline slightly.

Carl Bro International a|s and Associates

H- 4

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix H:

Traffic Growth Factors

Applying the forecast growth in motorcycle ownership to an estimate for 2000 of about 9
millions would give a total number of vehicles registered of about 15-20 millions in 2010 and
about 30-35 millions in 2020. This would give average ownership rates of about 12-15
motorcycles per 100 population in 2010, and 22-25 per 100 population in 2020. These are
high but probably fall short of saturation ownership. Similarly the car ownership figures
which result from the application of the forecast growth factors to an estimated ownership in
2000 of about 2.54 million would amount to about 5 million in 2010 and 10 million in 2020.
These are equivalent to about 4 per 100 population in 2010 and 7 per 100 in 2020, figures
which are still low by international standards and which fall well short of any likely saturation
values.
These figures are also broadly consistent with the findings derived from the cross-sectional
analysis of the PODES data, discussed in Technical Report No 9.
H3.2 Growth Factors by Province
In addition to the Java-wide figures, similar estimates have also been made at a more
disaggregated level, for East Java, Central Java/Yogyakarta and West Java/Jakarta. These are
set out in Tables F.5 - F.7. Similar broad conclusions to those discussed immediately above
apply.
Table H.5: Forecast Road Traffic Growth 2000,2010 and 2020, East Java
Vehicle Category
Motorcycles
Cars
LGV
Angkots
Buses
Trucks (MGV&HGV)
Total (All Vehs)

Traffic
Proportions
In 2000
44.6%
26.8%
6.9%
7.2%
2.3%
12.3%
100%

GRDP Growth

Carl Bro International a|s and Associates

Growth to
2010
1.72
1.72
1.38
1.05
1.33
1.38
1.60

Traffic
Proportions
in 2010
48%
29%
6%
5%
2%
11%
100%
1.52

Growth to
2020
3.09
3.09
2.18
1.10
2.04
2.18
2.75

Traffic
Proportions
in 2020
50%
30%
5%
3%
2%
10%
100%
2.52

H- 5

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix H:

Traffic Growth Factors

Table H.6: Forecast Road Traffic Growth 2000,2010 and 2020, Central Java &
Yogyakarta
Vehicle Category
Motorcycles
Cars
LGV
Angkots
Buses
Trucks (MGV&HGV)
Total (All Vehs)

Traffic
Proportions
in 2000
39.4%
26.5%
8.6%
7.5%
5.0%
13.0%
100%

Growth to
2010
1.86
1.86
1.51
1.05
1.46
1.51
1.70

GRDP Growth

Traffic
Proportions
in 2010
43%
29%
8%
5%
4%
12%
100%

Growth to
2020
2.95
2.95
2.10
1.08
1.98
2.10
2.56

Traffic
Proportions
in 2020
45%
30%
7%
3%
4%
11%
100%

1.48

2.37

Table H.7: Forecast Road Traffic Growth 2000,2010 and 2020, West Java & Jakarta
Vehicle Category
Motorcycles
Cars
LGV
Angkots
Buses
Trucks (MGV&HGV)
Total (All Vehs)

Traffic
Proportions
in 2000
19.4%
27.0%
10.4%
21.1%
3.3%
18.8%
100%

Growth to
2010
2.18
2.18
1.69
1.11
1.62
1.69
1.79

GRDP Growth

Traffic
Proportions
in 2010
24%
33%
10%
13%
3%
18%
100%

Growth to
2020
4.59
4.59
3.13
1.15
2.92
3.13
3.38

Traffic
Proportions
in 2020
26%
37%
10%
7%
3%
17%
100%

1.69

3.07

H3.3 Special Traffic Forecast Ferry Traffic

Forecasts of traffic growth on the ferry links to Sumatra (Merak-Bakauheni) and Bali
(Banyuwangi-Gilmanuk) have been estimated based on an analysis of traffic flows over the
recent years before the crisis. A simple linear relationship based on GRDP has been
estimated and is of the form:
Ferry traffic (Index 1993=1) = 1.27 * GRDP (Index 1993=1) -0.21 (R2 = 0.95)
Using this relationship, the growth factors, and the result are set out in Table H.8. These
factors have been applied to the base year traffic between Java and Sumatra and Java and
Bali, to give future year forecasts.
Table H.8: Ferry Traffic Growth Central GRDP Growth Scenario: (All Vehicle Types)
Ferry Between Java and:
Sumatra
Bali

Carl Bro International a|s and Associates

2000

2010
1.00
1.00

2020
1.72
1.71

3.05
3.09

H- 6

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix H:

Traffic Growth Factors

H4. GRDP OPTIMISTIC GROWTH SCENARIO TRAFFIC FORECASTS

H4.1 Optimistic Road Traffic Forecasts Java
Following the same procedures as described above for the optimistic GRDP growth scenario
the resultant traffic forecast growth factor for each region and by vehicle categories are given
below in Table H.9 and illustrated in Figure H.2.
Table H.9: Optimistic GRDP Growth Scenario - Traffic Growth Factors for 2010 and
2020
Vehicle Category
Motorcycles
Cars
LGV
Angkots
Buses
Trucks
(
)
GRDP Growth

West Java
& Jakarta

Cent. Java
& Yogyakarta

East Java

Java Total

20002010

20002020

20002010

20002020

20002010

20002020

20002010

20002020

2.80
2.80
2.06
1.12
1.95
2.06

7.72
7.73
5.01
1.28
4.61
5.01

1.88
1.88
1.45
1.06
1.39
1.45

4.26
4.26
2.88
1.01
2.68
2.88

2.15
2.15
1.63
1.08
1.55
1.63

4.89
4.89
3.24
1.06
3.00
3.24

2.44
2.44
1.82
1.09
1.73
1.82

6.58
6.58
4.31
1.14
3.98
4.31

2.05

4.89

1.72

3.30

1.84

3.83

1.93

4.33

Note: Index for GRDP and Vehicle Growth 2000=1.00

Figure H.2: Optimistic Growth Scenario Road Traffic Growth Forecasts Java

Optimistics Scenario Traffic Growth Forecasts

7.0
6.0
5.0
4.0
3.0
2.0
1.0
2000

2005
OPT-GRDP

Carl Bro International a|s and Associates

2010
Cars

2015
Trucks

2020
Buses

H- 7

Java Arterial Road Network Study (JARNS)

Technical Report No. 10, Appendix H:

Traffic Growth Factors

H4.2 Optimistic Ferry Traffic Forecasts

Using the same relationship as used above in Section F3.3, the growth factors for growth in
ferry traffic under the optimistic GRDP growth scenarios are listed below in Table H.10.
Table H.10: Ferry Traffic Growth Optimistic GRDP Growth Scenario: (All Vehicle
Types)
Ferry Between Java and:
Sumatra
Bali

Carl Bro International a|s and Associates

2000

2010
1.00
1.00

2020
1.97
1.92

4.29
4.17

H- 8