THE CHALLENGE OF CONTAINER TRANSSHIPMENT

IN THE CARIBBEAN
Ernst G. Frankel
Introduction
Increased use of mega containerships in transocean trades is forcing the development and use of
large transshipment hubs serving as container distribution and collection centers for large trading
areas. The economies of large-scale transshipment in mega container hubs in the Caribbean are
discussed, alternative feeder strategies are advanced and investigated, and the competitive
positions of alternative Caribbean transshipment hubs are evaluated.
Various feeder routing and scheduling strategies are investigated in terms of the impact on feeder
and hub port demands, import and export balances, mainline vessel size, speed, and service
frequency. Similarly, feeder vessel size and speed and the impact of inter-feeder and hub port
distances in determining feeder requirements are evaluated. The models include consideration of
transshipment hub port and feeder port container transfer (or ship turnaround) investment and
operating costs. An economic strategy tool designed to permit evaluation of the benefits of
transshipment in a mega transocean containership operation, with particular reference to
alternative transshipment hubs in the Caribbean, is presented.
General
The Caribbean, though small in population and economic activities except for tourism,
constitutes a major crossroads for international trade. This not only because of its proximity to
the Panama Canal, but also as a major focal point for north/south Atlantic trades and trades with
the east and northern coasts of South America. East coast South American economies have
rallied in recent years and are expected to continue their rapid growth notwithstanding recent
economic upheavals in Argentina. Container trade between North and South America has seen
the entrance of several new carriers in 2001 and is expected to grow by 3.9% in 2002. The
European-South American trade mostly with Brazil, Argentina, Venezuela, etc. is expected to
grow at a more modest 3% in 2002. South American population will continue to grow at twice
the rate of Europe and North America and its population is expected to exceed 700 million by
2025.
Notwithstanding the recent economic problems in Argentina, Uruguay and Brazil and problems
in Venezuela and its neighbors, South American economies are expected to revive and continue
their growth shortly. The major problem for lines serving Northern and Eastern South American
countries are the lengths of their coastlines and the resulting huge interport distances, combined
with significant but not large container trades, mostly foreign trades. This makes it difficult to
justify direct shipping by large containerships.

Similarly, the large distances to their principal trading partner ports make the use of smaller, less
efficient container ships unattractive. This has led to the development of transshipment ports in
the Caribbean designed to serve South America and the Gulf of Mexico.
The objectives of transshipment are not only to reduce the total cost of collecting and/or
distributing the containers carried by a mega-mainline container vessel from and to numerous
origin and destination ports, each of which only contributes a part of the mainline vessel cargo,
but also to improve just-in-time delivery of cargo, reduce in transit inventory, and make the total
origin-to-destination movement of containerized cargo more seamless. In other words, the
purpose is not just to reduce origin-to-destination transport and handling or transfer costs but to
make the whole supply chain, including all involved transactions, more efficient and more
responsive to the ever-changing market place.
Such demands include assurance of in-time delivery, reduction of technological risk of
obsolescence and technological competitiveness of traded products, customization of cargo at
transshipment points to meet special requirements, often communicated just before delivery, justin-time satisfaction of changing environmental and safety regulations, and other special currency
or payment and delivery requirements.
Transshipment is therefore not just a logistic convenience measure, but also an opportunity for
adding value to the goods transshipped and to the value of the logistic chain performance.
Consideration of the economics of transshipment must include all logistics as well as value
added activity costs and benefits contributed by transshipment activities.
Other incentives driving transshipment are the potential for reducing the impediments and costs
of cabotage by using foreign transshipment ports as a base for feeder transport to and from a
whole like of national ports. This is of particular importance when cabotage rules are imposed
on larger economic regions such as North America and in future the European Union and
elsewhere.
Transshipment also offers opportunities for cargo consolidation or deconsolidation and value
added activities such as assembly, calibration, and customizing to meet specific local or time
varying demands. To make transshipment attractive the economic and operational benefits must
outweigh added economic and operational costs such as additional handling costs, port dues, and
possibly extra voyage distances or deviations. At the same time, transshipment is often
necessary to attain economies of scale in shipping as well as the overall logistics chain.
During recent years transshipment has caught on in the Caribbean and a large number of
transshipment ports have been developed. Additional transshipment ports are under construction
or being planned. While the basic premise of transshipment is evident, the question arises if
radical changes in logistics or supply chain operations and technology introduce new challenges
which may negate some of the basic assumptions underlying the transshipment concept as
operationally and economically advantageous. In Asia, established mega transshipment ports
such a Singapore are under attack not just by adjacent Malaysian transshipment/gateway ports
such as Pelapas but smaller ports that used to serve largely as feeder ports. Similarly airlines that
invented transshipment hubs have largely abandoned the concept.

Table 1: Caribbean Transshipment Ports (2002)

U.S. South Atlantic Coast and Bahamas
Miami (US)
Port Everglades (US)
Jacksonville (US)
Savannah (US)
Charleston (US)
Freeport (Bahamas)
Caribbean
San Juan (Puerto Rico)
Colon (Panama)
Manzanillo (Panama)
Cartagena (Colombia)
Barranquilla (Colombia
Pt. Cabello (Venezuela)
Kingston (Jamaica)
Havana (Cuba)
Ponce (Puerto Rico)
Cristobal (Panama-Pac)
Balboa (Panama-Pac)
Rio Haina (Dominican Rep.)
Bridgetown (Barbados)
Pointe-a-Pietre (Guadeloupe)
Vieux Fort (St. Lucia)
Point Lisas (Trinidad)
The major problem of transshipment is that it introduces added transfer, storage, and transaction
costs and times which often exceed the cost and time saving introduced by faster, larger
container and feeder vessels.
Inter- and intramodal transfers which are usually still performed one container at a time
introduce major seams in the logistics chain where a mega container vessel may require
thousands of trucks to deliver and pick up a major part of or a complete shipload of containers.
Even when large numbers of container gantries are employed on a wharf, major obstacles
prevent smooth ship to/from truck or rail car container flows. As a result, a majority of
containers usually pass through buffers or stacks in mainline and even often in feeder
containership terminals.
More seamless intramodal transfer can usually be accomplished when supply chain links are
simplified through links and are added to the chain. This in addition to the potential reductions
in origin/destination times in supply chain or buffer inventories, near seamlessness, and

improvements in just-in-time delivery has made transshipment an element in intercontinental or

transocean container shipping.
Other advantages of transshipment are opportunities for rerouting and customizing of cargoes.
Large transshipment ports are major modal points in logistics management and offer the
opportunity for efficient transaction management. They offer opportunities for and the required
flexibility to respond to changes in supply chain structure, sequence, and direction imposed by
new demands or market conditions. Transshipment provides flexibility to supply chains and
therefore additional opportunities to logistics management. It also allows improvements in the
utilization and productivity of individual logistics links.
The Internet and wireless communications have made it possible to take real advantage of the
economic opportunities offered by transshipment by allowing multi-media, real-time
communications that simplify both operational as well as transaction management of logistic
services and operations.
Globalization of trade and outsourcing requires the ability to not only effectively manage logistic
chains but also change their structure or direction in near real time to respond to new technology,
market, and economic conditions. Outsourcing has become a global activity that requires large
flexibility in the dynamic matching of supply and demand.
Similarly, transshipment offers the opportunity for rapid changes in logistics chains. The number
of possible feeder strategies is usually very large. Feeder ports can be served one by one, in
circular, backtracking or other sequences, as will be discussed later in this paper.
Caribbean Container Transshipment Opportunities
Container traffic to and from South America is expected to grow from 13.628 m TEU in 1999 to
20.705 m TEU in 2005, as shown in Table 2. The Gulf of Mexico and the north and east coasts
of South America accounted for 8.621 m TEU in 1999. Over 4.5 m TEU or about 52% of this
trade benefits from transshipment (Table 3). These exclude U.S.-South American and intraSouth American trades which are assumed to benefit from direct shipping, as is U.S. Gulf and
east coast trade transshipped at Freeport, etc.
Although Freeport serves mainly transshipment of U.S. east coast cargo, it also offers major
advantages for South American trade with Europe, South, South East, and East Asia trade which
traverses the Mediterranean. Such trades are increasingly combined with U.S. East Coast/Gulf
Coast trades and carried by mega container ships. There is also a growing trend towards
Caribbean port transshipment of South American and U.S. East/Gulf Coast cargo by transpacific
vessels using the Panama Canal and large transAtlantic container ships carrying West Coast
North and South American cargoes which are transferred to Panamax container ships to facilitate
canal transit. Caribbean transshipment port capacity has doubled during the last 5 years (19972002) and now exceeds 6 million TEU/year, with another 2 million TEU/year capacity expected
to come on line by the end of 2003. The growth of excess capacity is expected to result in large-

scale competition among transshipment ports and feeders serving them with consequent rate
reductions.

TABLE 2: FORECAST GLOBAL CONTAINER PORT THROUGHPUT, 1996-2005 (000 TEU)

Region

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

19992005
Growth

North America

22,957

24,485

25,494

26,417

27,373

28,321

29,293

30,182

31,283

32,306

7,821

Western Europe

34,659

38,575

41,549

44,483

47,358

50,188

53,109

56,103

59,148

62,261

23,686

Far East

44,571

47,839

49,316

51,896

54,664

57,644

60,673

63,764

66,966

70,249

22,410

South East Asia

23,260

25,595

26,549

28,247

30,481

33,258

36,394

39,683

43,128

46,741

21,146

Middle East

7,392

8,004

8,546

9,258

10,066

10,890

11,670

12,473

13,292

14,136

6,132

South America

10,331

11,478

12,524

13,628

14,706

15,844

17,008

18,202

19,436

20,705

9,227

Australasia

3,592

3,794

3,939

4,096

4,255

4,408

4,565

4,728

4,888

5,055

1,261

South Asia

3,805

4,311

4,746

5,245

5,779

6,380

7,037

7,736

8,467

9,228

4,917

Africa

5,069

5,301

5,699

6,090

6,466

6,898

7,321

7,755

8,212

8,684

3,383

807

933

1,046

1,165

1,288

1,410

1,536

1,664

1,793

1,926

993

156,443

170,315

179,408

190,525

202,436

215,241

228,606

256,613

271,291

9,093

11,117

11,911

12,805

13,365

242,389
`
13,783

14,224

14,678

Eastern Europe
Total
Additional TEU

Source:

Drewry Shipping Consultants Ltd.

100,976

Table 3: Caribbean Container Trades and Transshipment Opportunities

Excluding transshipment of U.S. trade at Freeport, Bahamas, etc.

Transshipment Logistics Management

In transshipment, both fixed and variable strategies can be used. Variability may include
schedule, sequence, capacity, and more, and may be adjusted in response to changing market,
regulatory, user, and other conditions.
The transshipment port may be one or a range of ports (usually the one with the most traffic) or a
pure transshipment port without significant traffic of its own. There may also be more than one
transshipment port in a trading area. Usually each serves a distinct set of feeder ports, but that
may not always be the case and more than one transshipment ports may serve a particular feeder
port occasionally or all the time.
The number of possible feeder strategies is usually very large. Feeder ports can be served one by
one with a dedicated feeder in a two-way sequence, in a circular sequence, by pairs or flexibility
by volume. In the last case feeder vessels are assigned to one or more feeder ports until their
capacity is exhausted. In other words, we add additional ports to the feeder route until excess
capacity is largely used up.
Transshipment in the Caribbean is distinctly different from transshipment in most other ports of
the world. Usually mega containerships call at several transshipment ports in sequence (Hong
Kong and Singpore or East and West Mediterranean transshipment ports) which are served by
comparatively short feeder services. Furthermore transshipment ports generally generate
significant cargo themselves.
In the Caribbean transshipment ports, with the exception of San Juan (Puerto Rico), generate
insignificant volumes of cargo, and serve mostly very long feeders. In fact, Caribbean
transshipment ports often serve as the only or end of the line transshipment port. The economics
of transshipment in the Caribbean is as a result distinctly different as is the operational
management of transshipment. The problem is furthermore complicated when transshipment
also involves cabotage bypass (Freeport) and/or Panama Canal transit. Furthermore
transshipment in the Caribbean is often more complex than the typical transshipment models or
strategies described. Strategies must be dynamic as economic as market conditions are subject to
rapid and often unexpected change. Finally, unlike transshipment in other parts of the world,
Caribbean transshipment is nearly 100% ship-to-ship or intramodal with practically no
intermodal transshipment at all. As a result, logistic management in Caribbean transshipment
ports is distinctly different.
Economics of Caribbean Transshipment1
The economic impact of transshipment in the Caribbean is the sum of impacts such as:

2001.

In part based on paper presented at MPA Conference on Maritime Research organized by MPA in Singpore

Increased capacity and reduced cost of mainline shipping which makes only one or
two port calls at each end of the trade, thereby greatly reducing round trip time,
increasing frequency and capacity per unit time, and reducing cost per mainline slow.
Value of reduced origin to final destination delivery time, if any.
Added cost of transshipment in terms of unloading/reloading of containers as well as
dwell time costs.
Added cost of feeder service.
Added transactional costs, port costs, etc.

The economic impact is usually computed comparing the origin-to-destination costs and times of
direct mainline vessel pickup and delivery to those of mainline-transshipment-feeder vessel
pickup and delivery services. In each case alternative ship size, speed, routing, and scheduling
alternatives must be considered. Mainline ship size will usually indirectly determine service
frequency, though obviously demand is affected by service frequency and quality, the latter again
as a function of service frequency and time from origin to delivery. In other words, different
parameters in the system affect the outcome or the economics of the system and are
interdependent. For example, assuming no transshipment, a 4000 TEU mainline vessel would
offer about twice the frequency of an 8000 TEU vessel assuming the same routing, but a
difference in slot voyage costs between the smaller and larger mainline vessels. The question is
then if the increased frequency and therefore shorter average origin-destination time justifies
these higher costs by attracting added cargo. There are obviously other factors involved as well,
as some ports may not be able to accommodate very large mainline vessels. Transshipment
offers many more opportunities in terms of ship size, frequency, schedule, and routing of services
that affect overall cost and quality of service. The overriding factors in transshipment are added
unloading, loading, and transit costs, as well as related port wharfage, tonnage, and related
charges. On the other hand, transshipment offers opportunities for value added and improved
logistics management activities.
To analyze the economics of transshipment, we are concerned with the costs and delivery times
of direct versus transshipment service. Assuming the transshipment port is also a cargo hub, we
can determine the cost differential between direct delivery and transshipment service by
computing for each alternative strategy the sum of the costs of

mainline service to/from transshipment port

unloading and loading at the transshipment port
waiting at the transshipment port including stacking/unstacking, etc.
feeder transport between transshipment port and feeder ports
unloading and loading at the feeder ports

and compare it to the sum of the costs of direct service

mainline service to regions ports

unloading and loading at hub and feeder (regional) ports directly

Similarly the quality of service, which is our case is basically origin to destination delivery time,
can be computed for each alternative strategy by computing the different voyage, port, and cargo
waiting times.
Modeling Container Transshipment in the Caribbean
As noted before, unlike most transshipment activities, Caribbean container transshipment is
distinct due to

large feeder distances to/from south

significant U.S. American feeder and cabotage bypass trade
trans Panama Canal feedering
very small indigenous container trade
large variations in South American trade and resulting trade instabilities

Unlike other transshipment trade which usually serves more than one major transshipment hub at
each end of its route, Caribbean transport is often concentrated on one Caribbean transshipment
hub. To effectively model this situation we must divide the trade into its major components

East Coast South American trade

West Coast (Panama transit) South American trade
West Coast U.S. and transPacific (Panama transport) trade
U.S. East Coast and Gulf Coast trade
Local Caribbean/Gulf of Mexico trade

In each of these trades, circular, backtracking, and direct delivery feeder systems as well as
combinations of these alternatives can be used. The basic model for circular, backtracking, and
direct delivery feeder services is shown in Appendix A.
The alternatives to the use of a Caribbean transshipment hub are the use of direct mainline
delivery to East and West Coast of South America (using a circular, backtracking or direct port or
ports strategies) plus use of regional Caribbean transshipment (serving Caribbean and Gulf ports)
plus direct U.S. port delivery system and direct Panama transit system using panamax mainline
vessels.
Using 6000 TEU mainline vessels as an example and a typical transAtlantic route (NW Europe
to Caribbean) to serve U.S. East Coast, Gulf of Mexico, and East/West Coast South America via
one Caribbean transshipment ports plus direct service to each of these markets by smaller
mainline vessels, and/or use of several separate dedicated transshipment ports, one for each or
pairs of these major markets, we used a deterministic model constructed to simulate the times
and costs required to satisfy these markets for typical demand scenarios, using for each major
market (say East Coast South America)

direct return feedering of each feeder port separately with designated vessels or if
mainline vessel frequency large feeders may serve more than one feeder port
circular feedering of 2 or more feeder ports with ports called in circular sequence
without any backtracking
return feedering of 2 or more feeder ports using traditional backtracking from last
feeder port
backtracking of all feeder ports

and compare the results with direct non-transshipment service by vessels of 2000-3000 TEU,
using any of the above-named strategies. Similar vessels would be employed in the feeder
service.
Assuming, for example, a nominal Caribbean transshipment port (average transAtlantic distance
3200 nm) and feeder distances to 3 ports on East Coast of South America of 1400, 1800, and
2400 miles from transshipment port or 4200, 4600, and 5200 miles from North West Europe, the
following results were obtained. Demands were hypothetically assumed as 2000 TEU/week for
each of the feeder ports or 6000 TEU/week transshipped at the Caribbean transshipment port.
Assume interfeeder port distances of 400 and 600 miles, average mainline and feeder vessel
(2000-3000 TEU) speed of 21 knots, average mega vessel speed of 23 knots, average
loading/unloading rate at regional feeder ports of 50 moves/hour (including lost time, off time,
etc.), and 120 moves/hour at the transshipment port and assuming that on average each regional
port generates 1000 TEU/week of imports and 1000 TEU/week of exports or 1000 moves/week
(assuming all 40 ft. containers).
In both direct (mainline and feeder ship delivery), we can use the following strategies:

direct dedicated ships to each feeder port

direct return shipping and feedering
circular path shipping and feedering
backtracking shipping and feedering

Table 3 shows the interport distances and travel times.

Ports
Europe origin
Caribbean
transshipment
Port A
Port B
Port C

Table 3: Interport Distances and Travel Times (hours)

Europe Origin Caribbean
Port A
Port C
Transhipment
-3200 (139)
4200 (200) 4600 (219)
3200 (139)
-1400 (67)
1800 (86)
4200 (200)
4600 (219)
5200 (248)

1400 (67)
800 (86)
2400 (114)

-400 (19)
1000 (48)

400 (19)
-600 (29)

Port C
5200 (248)
2400 (114)
1000 (48)
600 (29)
--

Similarly port times are calculated assuming the productivity moves/hour per berth noted before
as shown in Table 4. These include a 4-hour docking, mooring, and preparation time.
Berth Moves/Hr
Europe
Transshipment *
Ports A, B, C
each

80-120
80-120
50

Table 4: Port Times

Direct Mainline
Mega Vessel
Service 4000
TEU Vessel
41.50
54.00
-54.00
Direct 64.00
-Circular 2400
Backtracking
24.00

Feeder Vessel
3000 TEU
-41.50
Direct 64.00
Indirect 24.00

Berth moves/hour mega vessel 120/hour 3000 TEU vessel 80/hour

Direct service to Ports A, B, and C imply designated mainline or feeder vessel of 3000 TEU
(3000 moves per port call).
The resulting round-trip times are as follows:
Transshipment Alternatives
Mega Containership (6000 TEU)
Europe-Caribbean transshipment port
Direct Feeders
Caribbean Port A
Caribbean Port B
Caribbean Port C
Circular Feeders (Caribbean-A-B-C-Caribbean)
Backtracking Feeders (Caribbean-A-B-CB-ACaribbean)
Direct Shipping (without transshipment)
Designated: Direct Europe to each Port A
Direct Europe to each Port B
Direct Europe to each Port C
Direct Europe Circular
Direct Europe Backtracking

386.1 hours
239.5 hours
277.5 hours
333.5 hours
342.5 hours
368.0 hours
505.5 hours
543.5 hours
601.5 hours
609.5 hours
621.0 hours

The operating costs of mega mainline vessels are estimated to be $2000/hour at sea and
$1200/hour in port for direct (2000-3000 TEU) mainline or feeder vessels $1390/hour at sea and
$860/hour in port. As a result, round-trip ship costs for each alternative considered can be
computed, as well as the origin (Europe) to destination (Ports A, B or C) time for direct delivery
as well as transshipped delivery. Consideration must be given to transshipment delays resulting
from mega and feeder ship schedule differences. These vary from an average of 25 to over 72

hours. Using the round-trip times of the direct and transshipment service strategies, the total
number of vessels (mega, mainline, feeder) required to meet the demands projected can be
determined and priced to establish the cost per TEU of providing the shipping (transport)
services. Similarly, the average (minimum and maximum) origin-to-destination times achievable
using alternative strategies can be determined.
In addition, the costs of cargo loading and unloading as well as other port related costs must be
added. While different direct delivery and feeder ship strategies introduce some small
differences in origin (Europe) and destination direct port (A, B, C) costs, the major difference is
really the added port costs and time introduced by transshipment. This can be significant and
may add several hundred dollars per container or about $100-150/TEU. This in turn can readily
exceed any savings in shipping costs offered by the efficient use of mega container vessels.
Obviously the routing and scheduling alternatives to serve a particular trading area must
consider:

different potential ship capacities, speeds, and resulting frequency of service to meet
demand
scheduling of mainline and feeder ships
routing of mainline and feeder vessels
route specific requirements such as working days per week, special customer service
demands, port dues, and container handling costs, etc.
special backhaul requirements, if any

To handle these, the use of implicit enumeration algorithms was found to be useful.
In our simple example the cost and time of direct delivery or shipping costs (excluding European
and Ports A, B, and C port costs) would be assuming 100% full TEU both ways.
Direct Delivery Port A
Direct Delivery Port B
Direct Delivery Port C
Direct Delivery Circular
Direct Delivery Backtracking

Total
$625,730
$679,550
$760,170
$787,050
$798,400

Cost/TEU
$104.46
$113.26
$120.67
$131.17
$133.07

Costs with transshipment excluding transshipment port costs and other port costs (each mega
ship serves two feeders).
1 mega ships Europe to
Caribbean transship
Direct feeder Caribbean to A
Direct Caribbean to B
Direct Caribbean to C
Total cost of transship with

Total
$960,000

Cost/TEU
$53.33

$266,900
$319,810
$400,432
$1,940,600

$46.15
$54.96
$68.40
$107.81

direct feeder of one mega ship

transferred to three feeder
vessels
Obviously direct delivery between the transshipment port and the South American ports may not
be the most economic approach. At the same time the cost savings compared to Europe-South
America direct delivery with average shipping costs of $114.8 is only about 7% more expensive,
a margin which is totally inadequate to justify (or pay for) the added transshipment handling
costs.

The above example assumes full loads in both directions on mega, mainline, as well as feeder
vessels. Obviously, this example is unrealistic and therefore the model was used to determine
the economics of transshipment for a range of ship sizes (mega, mainline, and feeder vessels),
ratios of distances between origin port(s) and transshipment port(s) and transshipment port(s) and
feeder port(s), as well as traffic volumes.
All the computations attempted to determine total costs (origin and destination ports as well as
all shipping costs) so as to identify the gap available to cover the costs of transshipment
(transshipment port and related costs). Using the basic example presented before and moving the
transshipment port location along the Europe-South America route, computations readily showed
that the ratio of mega to average feeder ship distance affect the percentage savings as shown in
Figure 2.

Figure 2: Savings of Shipping Costs as a Function of Transshipment Port Location

Example:

6000 TEU Mega Ship

3000 TEU Mainline Vessel
3000 TEU Feeder Vessel

Similarly, computations were made for the whole range of transshipment port locations, O-D
distances, ship sizes, and trading volumes. They distinctly show that transshipment effectiveness
is highly dependent on distance ratios, overall distances, relative ship sizes, trading volumes, and
most importantly costs of transshipment.
Possible Structural Representation
Converging on a solution to the ship routing, scheduling, sizing, and speed selection alternative
in transshipment problems can be achieved by algorithms based on implicit enumeration using
branch and bound solution techniques. The procedure only schedules feasible solutions. At each
step of the bounding process, alternative routes and associated schedules are introduced.
Ultimately the branching procedure generates sets of feasible solutions which are subjected to a
trade-off between origin-destination times and total shipping, including transshipping, costs.
The value of time is difficult to estimate because of the large variety of cargoes shipped in
containers. While high value cargoes put a premium on time and their shippers are willing to
pay more for faster delivery, others may be less inclined. As a result, transshipment system
designs may have to differentiate between different cargoes and their time value. This may
permit a role for both limited, high speed feeder systems and very low cost, low speed feeders. It
may also allow the introduction of RoRo high sped feeder ferries in combination with mainline
LoLo vessels. This particularly when feeder distances are comparatively short. For example,
RoRo feeders are very economic for feeder distances of 100-500 nm, particularly on routes with
inadequate feeder port LoLo container terminals. As a result, transshipment terminals may have
to offer alternative feeder berth facilities, particularly if RoRo feedering offers lower total
logistics costs and delivery times to or from some destinations. In future large transshipment
activities may follow the lead of world class courier services which differentiate between
different cargoes. It really makes little sense to provide the same service for very low value, time
sensitive cargoes than for very high value, highly time sensitive cargoes.
In our analysis, these factors assume important considerations. Origin-destination time
incentives are assumed to attract more high value freight. Because high speed, short distance
feeder transport often has significantly faster turnaround times than mainline inter-arrival times
at the transshipment port, various high speed feeder routings or multiple service uses may be
appropriate.
Structuring a mainline transshipment port-feeder services logistics problem therefore requires
consideration of many

service frequency alternatives

scheduling alternatives
routing alternatives
feeder size alternatives
feeder speed alternatives
feeder technology alternatives

feeder terminal conditions

It similarly requires knowledge of the existing market demand and potential demand for higher
speed/quality service as well as the potential premium value of time savings.
Another important issue is service frequency that in some trades is important as origindestination times, particularly when continuous demand permits much lower inventory levels
with high service frequency. This obviously also affects transport and in transit inventory. For
high value, time sensitive cargoes this can be as important as time to delivery.
It is usually assumed that cargo with a value per TEU of $100,000 has a time value of $100-200
per day which includes real cost of time, holding, as well as costs of potential obsolescence or
market acceptance decline. In other words, a day saved in the delivery of 1000 TEU of high
value cargo is worth $100,000-$200,000. This must be considered in trading off frequency and
time to delivery.
Modeling Approaches
Network and implicit enumeration algorithmic models have been used to represent the complex
multiple routing, scheduling, feeder size, speed, and technology transshipment problem, and
solved as queuing or stochastic queuing methods. Linear programming branch and bound
techniques are useful in solving implicit enumeration models. In reality there are obviously
significant statistical deviations in all the factors or parameters involved in transshipment.
In is important to incorporate and determine the conditional probabilities that a certain port or
ship service will be used and given it is used, the probability that it will perform is assumed.
Conditional stochastic queuing network algorithms are found to offer effective methods for
modeling conditional probabilistic, stochastic logistic networks representing mainlinetransshipment port-feeder line operations. The results obtained are

total costs and distribution of costs of providing the services

the risk in providing the service in terms of revenue generation or probability
statistics of O-D times for all and/or groups of different designated cargoes
the expansion capability of the service

Conclusions
Transshipment in container shipping logistics offers greater flexibility, potential cost savings, as
well as potential improvements in O-D times. But to be effective, transshipment must be
effectively designed. Most importantly, it must be highly efficient. In many cases the cost of
transshipment is not justified by savings in shipping costs. Often transshipment cost must be
brought down to as low as $40 per container to justify transshipment.

Transshipment, particularly in the Caribbean, with little if any indigenous cargo, requires very
effective analysis, efficient intramodal low cost transfer, and value added activities to be
successful. Value added activities can consist of physical improvements (customization,
assembly, etc.) of cargo, cabotage bypass, and various other operations. Transshipment based
solely on the assumed benefits of economies of scale of mega ships interfacing with smaller
feeder vessels may not provide real benefits.

APPENDIX A: Basic Transshipment Feeder Service Model

Assuming a simple example of a hub which can also serve as a transshipment port and just two additional regional
(feeder) ports, let

d 02
dOH
dH1
dH 2
d12

= distance between last feeder and overseas port

= distance between overseas and hub ports
= distance between hub and feeder port 1
= distance between hub and feeder port 2
= distance between feeder ports 1and 2
(Note: all in nautical miles)

If the mainline vessel has a voyage speed of

are:
a.

vm knots, then the total round trip voyage times of a mainline vessel

Circular Routing Strategy

TVI dOH d H 1 d H 2 d12 d02 / vm

b.

Backtracking Strategy

TVII 2 dOH d HI d12 / vm

Adding port time given:

WH , W1 , W2 are respectively amount of cargo (no. of lifts) destined to and WHI , WIII and W21 (no. of lifts
originating from the hub and the two regional feeder ports then given W21 handling rates in lifts per hour are hH ,
h1 , and h2 and entry/exit time loss is nH , n1 , and n2 then the total time in port spent using strategy (a) circular
routing is

TpI WH WH1 / hH nH W1 W11 /

h1 n1 W2 W21 / h2 n2

Similarly for backtracking operations

TpII TpI nH n1 n2
Therefore total round trip times for the Circular and Backtracking Routing Strategies are

T I TVI Tp1
and

T II TVII TpII
Similarly assuming transshipment, the mainline vessel round trip time is

TV [2dOH / vm WH WHI W1 W12 W2 W21 /

hH nH ]

This paper is part of the

IAME Panama 2002 Conference Proceedings

The paper has been anonymously peer reviewed and accepted for presentation by the
IAME Panama 2002 International Steering Committee
The conference was held on
13 15 November 2002
in Panama
The complete conference proceedings are published in electronic format under
http://www.eclac.cl/Transporte/perfil/iame_papers/papers.asp
For further information contact [email protected]