Introductory Elevator Traffic System Design (METE VI)

Lutfi Al-Sharif
Mechatronics Engineering Department, School of Engineering
The University of Jordan, Amman 11942, Jordan

Revision 4, 10th June 2016

Nomenclature
τ is the round trip time in s
a the rated acceleration in m/s2
AR% the passenger arrivals expressed as a percentage of the building population
arriving in the busiest five minutes
CC the nominal car carrying capacity in persons
df the height of a floor in m
df(i) the height of the ith floor in m
dG is the height of the ground in m where more than the typical floor height
H is the highest reversal floor (where floors are numbered 0, 1, 2….N)
HC% the handling capacity expressed as a percentage of the building population that
can be transported in five minutes
int the nominal interval in s
intact the actual interval in s
j the rated jerk in m/s3
L the number of the elevators in the group
N the total number of floors above the main entrance
P is the nominal number of passengers in the car when it leaves the ground floor
(can be a fraction)
Pact is the actual number of passengers in the car when it leaves the ground floor
(can be a fraction)
Parr(i) the percentage arrival from an entrance floor
S is the probable number of stops in a round trip
tao the door advance opening time in s (where the door starts opening before the car
comes to a complete standstill)
tdc the door closing time in s
tdo the door opening time in s
tf the time taken to complete a one floor journey in s
tpi the passenger boarding time in s
tpo the passenger alighting time in s
tsd the motor start delay in s
U the total building population
Ui the building population on the ith floor
v the rated speed in m/s

As a general rule, lowercase symbols represent random variable and uppercase

symbols represent expected (average) values (constants).

1.0 INTRODUCTION
Part I of this series of articles introduced the general concept of demand [1]. Parts II,
IV and V of this series of articles ([2], [4] and [5]) have introduced the concept of the
round trip time and different methods for calculating it. Calculating the round trip time
is the building block of elevator traffic analysis and design. Rather than being an
objective in itself, the evaluation of the round trip is a tool in order to carry out the
overall design of the elevator traffic system. More details about the evaluation of the
round trip time can be found in ([10], [12], [13], [15] and [17]).
This section will provide an introduction to overall approach to design. It will
also follow a staged approach to the problem of elevator traffic design through the
use of a series of carefully selected examples that progressively increase in
complexity in order to clarify specific concepts.

2.0 GENERALISED DESIGN APPROACH

Any compliant design must meet the two user requirements: quality of service as
represented by target interval (INTtar) and the quantity of service as represented by
the arrival rate (AR%) ([8], [9]). Having introduced the method for calculating the
round trip time, it is now possible to introduce the general approach for finding the
required number of elevators, L. It is assumed that the L elevators will operate in one
group. Under conventional group control, any passenger requesting service can
board the first arriving elevator of any of the L elevators in the group.
Once the round trip time has been evaluated, dividing it by the target interval
stipulated by the user will provide the number of required elevators in the group, L, as
shown in equation (1) below:

 RTT 
L=  (1)
 INTtar 

The x  operator is referred to as the ceiling operator and is defined as the smallest
integer not less than x. It is in effect a “rounding up” operator. It recognises the fact
that a whole number of elevator cars must be selected. From this value of L, it is
possible to find the actual interval that will be achieved in practice.

RTT
INTact = (2)
L

And the handling capacity can be calculated as follows:

300 ⋅ P ⋅ L
HC % = (3)
τ ⋅U

There is a no guarantee the quantity of service requirement will be achieved. In

other words the inequality shown below is not necessarily true:

HC % ≥ AR% (4)

It is important to note the effect of the assumed car capacity on the solution. The
selection of the car capacity (CC) will determine the value of the average number of
passengers, P, boarding the elevator car in each round trip. The number of
passengers, P, will in turn affect the value of the round trip time and hence the
required number of elevators in the group, L.
 It is assumed that the designer will start with a pre-defined car capacity (CC)
in units of passengers. As has been discussed in the earlier chapters, the car will not
fill up to its full capacity and it is customary to assume that it will only fill up to 80% of
its rated capacity.
Setting the value of P as equal to 80% of the rated car capacity (CC) could
result in a fractional number. This is acceptable and is caused by the fact that the
number of passengers P boarding the car is an average of all the various round trips.
In fact, P is a random variable and the calculations presented here are using its
average value.

3.0 DESIGN EXAMPLES

This section presents two simple design examples that will provide a simple
methodology for elevator traffic design. The two examples assume the following
conditions in the building:

1. Equal floor heights.

2. Top speed attained in one floor journey.
3. Single Entrance.
4. Up peak traffic only.

Example 1: Equal floor populations

A building has 6 floors above the main entrance, a single entrance. The elevator
cars to be used have a rated capacity of 13 persons. The building has a total
population of 960 persons and equal floor populations (160 persons per floor).
Assume the following parameters:
Floor to floor height: 3.5 m
Rated speed: 1 m/s
Rated acceleration: 1 m/s2
Rated jerk: 1 m/s3
Passenger boarding/alighting time: 1.2 s
Door opening time: 1.8 seconds.
Door closing time: 2.7 seconds.

The user requirements are:

Target interval of 30 s.
Arrival rate is 12% of the building population in five minutes.

It will be assumed that the car will fill up to 80% of its rated capacity. So it will be
assumed that the number of passengers used in the round trip time equation is 10.4
passsengers. The fact that this number is not an integer need not be a problem as it
represents an average over a number of journeys.
As the floor populations are equal, the formula for the highest reversal floor
can be used to find the average value of the highest reversal floor, H:

N −1 i P 
H = N − ∑   (5)
 
i =1   N  

Substituting the value of N and P produce the value of the highest reversal floor (in
units of floors):
 5   i 10.4 
H = 6 − ∑     = 5.834 (6)

i =1   6 



In a similar way, the expected number of stops in a round trip can be evaluated by
applying the formula for S under the assumption of equal floor populations:

  1  
P

S = N ⋅ 1 − 1 −  (7)
  N 
 

Substituting the value of N and P produce the value of the expected number of stops
in a round trip (in units of stops):

  1 10.4 
S = 6 ⋅ 1 − 1 −   = 5.099 (8)
  6 
 

(N.B.: Carry out a check to ensure that the rated speed is attained in one floor
a2 ⋅ v + v2 ⋅ j
journey using the equation d f ≥ .)
a⋅ j

Using the equation for calculating the round trip time, based on the assumption of top
speed attained in one floor journey, equal floor heights and a single entrance under
incoming traffic conditions gives:

+ (5.099 + 1) ⋅ (2 + 1.8 + 2.7 ) + 2 ⋅ 10.4 ⋅ 1.2

3.5
RTT = 2 ⋅ 5.834 ⋅
1 (9)
= 105.44 s

Dividing this value of the round trip time by the target interval and rounding up, gives
the required number of elevators:

 RTT  105.44 
L= = =4 (10)
 INTtar   30 

Thus four elevators are required in order to meet the required target interval. The
actual interval will be:

RTT 105.44
INTact = = = 26.36 s < 30 s = INTtar (11)
L 4

And the handling capacity of the resulting system will be:

300 ⋅ P ⋅ L 300 ⋅ 10.4 ⋅ 4

HC % = = = 12.33% > AR% = 12% (12)
RTT ⋅ U 105.44 ⋅ 960
As the actual interval is smaller than the target interval and the handing capacity is
larger than the arrival rate, both of the design criteria have been met and thus the
system design is compliant. It is also worth nothing that the design is not wasteful as
the actual interval is not much smaller than the target interval and the handling
capacity is not much larger than the arrival rate.

Example 2: Equal floor populations

The following example also applies to a simple building and attempts to illustrate the
same concepts as those presented in example 1, but using a slightly larger building,
a higher arrival rate and a higher elevator rated speed.
A building has 10 floors above the main entrance, a single entrance. The elevator
cars to be used have a rated capacity of 16 persons. The building has a total
population of 800 persons and equal floor populations. Assume the following
parameters:

Floor to floor height: 4.5 m

Rated speed: 1.6 m/s
Rated acceleration: 1 m/s2
Rated jerk: 1 m/s3
Passenger boarding/alighting time: 1.2 s
Door opening time: 2 seconds
Door closing time: 3 seconds

Target interval of 30 s
Arrival rate of 15% of the building population in five minutes

It will be assumed that the car will fill up to 80% of its rated capacity, giving an
effective average number of passengers, P, in a round trip of 12.8 passengers.

(N.B.: Carry out a check to ensure that the rated speed is attained in one floor
a2 ⋅ v + v2 ⋅ j
journey using the equation d f ≥ .)
a⋅ j

As the floor populations are equal, the formula for the highest reversal floor can be
used to find the average value of the highest reversal floor, H (which has units of
floors):

9   i 12.8 
H = 10 − ∑     = 9.67 (13)

i =1   10 



In a similar way, the expected number of stops in a round trip can be evaluated by
applying the formula for S under the assumption of equal floor populations, giving the
average number of stops in a round trip:

  1  
12.8

S = 10 ⋅ 1 − 1 −  = 7.4 (14)
  10  
 
Using the equation for calculating the round trip time, based on the assumption of top
speed attained in one floor journey, equal floor heights and a single entrance under
incoming traffic conditions gives:

+ (7.4 + 1) ⋅ (2.6 + 2 + 3) + 2 ⋅ 12.8 ⋅ 1.2

4.5
RTT = 2 ⋅ 9.67 ⋅
1.6 (15)
= 148.96 s

Dividing this value of the round trip time by the target interval and rounding up, gives
the required number of elevators:

 RTT  148.96 
L= = =5 (16)
 INTtar   30 

Thus five elevators are required in order to meet the required target interval. The
actual interval will be:

RTT 148.96
INTact = = = 29.79 s < 30 s = INTtar (17)
L 5

And the handling capacity of the resulting system will be:

300 ⋅ P ⋅ L 300 ⋅12.8 ⋅ 5

HC % = = = 16.11% > AR% = 15% (18)
RTT ⋅ U 148.96 ⋅ 800

As the two conditions for quality of service (i.e., the interval) and the quantity of
service (i.e., the handling capacity) have both been met, the design is a compliant
design, but still not a wasteful design.

Example 3: Unequal floor populations

This example is an application of the case where the floor populations are unequal.
Different formulae for the value of H and S have to be used.
A building has 8 floors above the main entrance, a building population of 600
people and equal floor heights with a floor height of 4.2 m (df). The elevator traffic
system is to be designed for an expected arrival rate (AR%) of 12.5%. The elevator
cars to be used are rated at 10 persons (CC=10). The building population is
distributed as shown below:

Floor Population Floor Population

1 100 5 80
2 100 6 60
3 80 7 50
4 80 8 50

Select a suitable number of elevators in order to meet a target interval of 30 seconds.

Assume the following parameters:
• Rated speed 1 m/s
• Rated acceleration 1 m/s2
 • Rated jerk 1 m/s3
• Door opening time 2 s
• Door closing time 3 s

The first step is to calculate the values of H and S, using the formulae for unequal
floor population shown below:

N  Ui P 
S = N − ∑  1 −   (19)

i =1   U  

 j U
N −1

P

H = N − ∑   ∑  i    (20)

j =1  i =1  U  
 

It is easier to do this using a table, as shown below.

P
Ui  Ui 
P i
U j   i U j 
Floor (i) Ui 1 −  ∑   ∑ 
 j =1  U  
U  U  j =1  U    
1 100 1/6 0.2326 1/6 5.954x10-7
2 100 1/6 0.2326 2/6 1.52x10-4
3 80 2/15 0.3183 7/15 2.25x10-3
4 80 2/15 0.3183 3/5 0.01679
5 80 2/15 0.3183 11/15 0.08364
6 60 1/10 0.4304 5/6 0.2325
7 50 1/12 0.4985 11/12 0.4985
8 50 1/12 0.4985 N/A N/A
Sum 600 1 2.8475 0.83386

(N.B.: Carry out a check to ensure that the rated speed is attained in one floor
a2 ⋅ v + v2 ⋅ j
journey using the equation d f ≥ .)
a⋅ j

Using the results from the table above, the value of H and S are 7.1661 floors and
5.1525 stops per round trip, respectively. Substituting the relevant values in the
equation for the round trip time gives a value for the round trip time of 103.581 s.
Four elevators will be needed in order to meet the target interval of 30 s, as shown
below.

 RTT  103.581
L= = =4 (21)
 INTtar   30 

The actual interval is met as shown below:

RTT 103.581
INTact = = = 25.9 s < 30 s = INTtar (22)
L 4
.... as well as the handling capacity:

300 ⋅ P ⋅ L 300 ⋅ 8 ⋅ 4
HC % = = = 15.44% > AR% = 12.5% (23)
RTT ⋅ U 103.581 ⋅ 600

It can be noted that the handling capacity is significantly larger than the arrival rate,
and there is potential to reduce the number of passenger and thus the car loading.

In all of three previous examples, the resulting design was compliant. However, this
is not necessarily always the cases. In order to illustrate such a problem to the
reader, the following two examples result in wasteful or non-compliant designs. The
first example (example 4) shows a wasteful design, where the target interval has
been met but the handling capacity is excessive. The second example (example 5)
shows a non-compliant design that fails to meet both quantity of service and quality
of service.

Example 4: Large Handling Capacity and a compliant interval

A building has 10 floors above the main terminal, has an arrival rate of 15% of the
building population in the busiest five minutes, a total building population of 1000
persons equally distributed on the 10 floors. The floor heights are equal at a value of
4.3 m. Use a rated car load of 26 persons. The target interval is 30 s.

Assume the following parameters:

• Rated speed 1.6 m/s
• Rated acceleration 1 m/s2
• Rated jerk 1 m/s3
• Door opening time 2 s
• Door closing time 3 s

The number of passengers, P, will be calculated as 80% of the rated capacity, CC.
This gives a value of P of 20.8 persons (this is accepted as a fraction as it represents
an average value over a number of journeys).
With this value of P, and using the equations for H and S (assuming equal floor
populations), gives the following values for H and S:

9   i  20.8 
H = 10 − ∑     = 9.877 (24)

i =1   10 



In a similar way, the expected number of stops in a round trip can be evaluated by
applying the formula for S under the assumption of equal floor populations, giving the
average number of stops in a round trip:

  1 
20.8

S = 10 ⋅ 1 − 1 −   = 8.8825 (25)
  10  
 
(N.B.: Carry out a check to ensure that the rated speed is attained in one floor
a2 ⋅ v + v2 ⋅ j
journey using the equation d f ≥ .)
a⋅ j

The round trip time is then evaluated as shown below:

+ (8.8825 + 1) ⋅ (2.6 + 2 + 3) + 2 ⋅ 20.8 ⋅1.2

4.3
RTT = 2 ⋅ 9.877 ⋅
1.6 (26)
= 178.117 s

Dividing this value of the round trip time by the target interval and rounding up, gives
the required number of elevators:

 RTT  178.117 
L= = =6 (27)
 INTtar   30 

Thus six elevators are required in order to meet the required target interval. The
actual interval will be:

RTT 178.117
INTact = = = 29.686 s < 30 s = INTtar (28)
L 6

And the handling capacity of the resulting system will be:

300 ⋅ P ⋅ L 300 ⋅ 20.8 ⋅ 6

HC % = = = 21.02% >> AR% = 15% (29)
RTT ⋅ U 178.117 ⋅1000

Although the design is compliant, it is in fact wasteful, because the handling capacity
is much larger than arrival rate. The actual interval is slightly smaller than the target
interval and thus the system meets the quality of service criterion. Steps have to be
taken in order to reduce the handling capacity to reduce the waste in the design.

Example 5: Non-compliant Design (fails on both quality and quantity criteria)

A building has 7 floors above the main entrance. The number of elevators to be used
in the building has been fixed by the architect to two elevators, with a rated speed of
1.6 m/s and rated capacity of 10 persons. The floors heights are equal at a value of
4.5 m. The total building population is 560 persons, and the arrival rate (AR%) is
13% of the building population in the busiest five minutes. The target interval is 30 s.
Evaluate the actual interval and the handling capacity of the elevator system
and establish whether it is a compliant design.

Assume the following parameters:

• Rated speed 1.6 m/s
• Rated acceleration 1 m/s2
• Rated jerk 1 m/s3
• Door opening time 2 s
• Door closing time 3 s
The number of passengers, P, will be 80% of the rated capacity which is 8 persons.
Using this value of P, and assuming equal floor populations, the highest reversal floor
is:

6   i 8 
H = 7 − ∑     = 6.6284 (30)

i =1   7  


And the average number of stops per round trip, S:

  1 8 
S = 7 ⋅ 1 − 1 −   = 4.96 (31)
  7 
 

(N.B.: Carry out a check to ensure that the rated speed is attained in one floor
a2 ⋅ v + v2 ⋅ j
journey using the equation d f ≥ .)
a⋅ j

The round trip is calculated as shown below:

+ (4.96 + 1) ⋅ (2.6 + 2 + 3) + 2 ⋅ 8 ⋅1.2

4.5
RTT = 2 ⋅ 6.6284 ⋅
1.6 (32)
= 101.781 s

In this example, the number of elevators has been set to two elevators. Thus it is
possible to find the actual interval as shown below:

RTT 101.781
INTact = = = 50.8905 s >> 30 s = INTtar (33)
L 2

So the system fails to meet the quality of service criterion, as the actual interval is
much larger than the target interval. And the handling capacity of the system will be:

300 ⋅ P ⋅ L 300 ⋅ 8 ⋅ 2
HC % = = = 8.42% << AR% = 13% (34)
RTT ⋅ U 101.781 ⋅ 560

And the system also fails the quantity of service criterion, as the handling capacity is
much smaller than the expected arrival rate. Addressing this problem will have to
start by increasing the number of elevators.

4.0 THE DIFFERENT OPTIONS AVAILABLE TO THE DESIGNER

An illustrative example is presented in this section to show the large number of
different options available to the designer, in order to arrive at a compliant design.

Example 6: An illustrative Example

Let us take a simple building as an example to illustrate the methodology presented
above. The parameters of the building are shown below. It is assumed to have a
single entrance, equal floor heights and equal floor populations to simplify the
calculations. The calculation is obviously based on incoming traffic only.

User requirements
Target interval (inttar): 30 seconds
Arrival rate (AR%): 12% of the building population in five minutes

Building and elevator parameters

U, total population = 600 persons
N, number of floors above the main entrance = 8 floors

Door times
tdo= 2 s; tdc= 3 s

Passenger transfer times

tpi= 1.2 s; tpo=1.2 s.

Kinematics
a= 1 m·s-1; j=1 m·s-2

The total travel distance between terminal floors is

d tot = d f ⋅ N = 4.2 ⋅ 8 = 33.6 m (35)

Using a speed of 1 m/s will achieve a transit time between terminal floors of 35.6 s,
which is acceptable.

dtot v a 33.6 1 1
ttransit = + + = + + = 35.6 s (36)
v a j 1 1 1

Let us assume a car capacity of 1000 kg/13 persons. Using the 80% loading factor,
gives the number of passengers, P, equal to 10.4 passengers.
The next step is to find the value of the round trip time at the value of P = 10.4
passengers and a rated speed of 1 m/s. This produces a value for the round trip time
of 138.61 s. The number of required elevators can be found by dividing this number
by the target interval.

 RTT  138.61
L= =  = 4.62 = 5 elevators (37)
 INTtar   30 

The actual interval can be found as shown below:

RTT 138.61
INTact = = = 27.72 s (38)
L 5

The handling capacity can be calculated using equation (3) as shown below:
 300 ⋅ P ⋅ L 300 ⋅ 10.4 ⋅ 5
HC % = = = 18.8% (39)
RTT ⋅ U 138.61 ⋅ 600

But this value of handling capacity is much larger than the user requirement for
arrival rate (AR%=12%). So this solution has produced an interval that is smaller
than the target interval, but a handling capacity that is much larger than the arrival
rate (AR%).
The obvious next step is to reduce the car capacity. Let us use a car capacity
of 800 kg/ 10 persons (CC). Using the 80% ratio, results in a number of passengers
(P) of 8 persons. Using this value of P, we can evaluate the round trip time:

RTT (P = 8) = 126.2 s (40)

Using equation (1) to find the required number of elevators shows that five elevators
are required. The actual interval and the handling capacity are calculated again
below:

RTT 126.2
INTact = = = 25.24 s (41)
L 5

The handling capacity can be calculated using equation (3) as shown below:

300 ⋅ 8 ⋅ 5
HC % = = 15.85% (42)
126.2 ⋅ 600

This is a better solution as the handling capacity is nearer to the specified arrival rate
(AR%). Keeping the same car capacity (800 kg/10 person) we can further reduce the
number of passengers down to 7 persons. This will results in a value for the round
trip time of 120.2 s, an actual interval of 24 s and a handling capacity of 14.6% still
using five elevators. A further reduction in the number of passengers down to 5.1
passengers while keeping the number of elevators at five elevators will result in an
interval of 21.4 s at a handling capacity of 12%.
However, the design is wasteful, and it is possible to reduce the number of
elevators to four elevators. In fact, using trial and error it can be shown that using
four elevators with P = 7.3 passengers will produce a handling capacity of exactly
12% and an actual interval that is only slightly larger than the target interval (30.5 s).
This is in fact a reasonable solution and results in the reduction of the number
elevators in the group from five down to four elevators.

Design stage P L intact HC% Comments

1 10.4 5 27.72 s 18.8% Excessive handling capacity
2 8 5 25.24 s 15.85%
3 7 5 24.0 s 14.6%
Interval must smaller than
4 5.1 5 21.4 s 12%
the target interval
Acceptable design; just
5 7.3 4 30.5 s 12%
misses the target interval
From this example, it can be seen that in some cases, it is possible to reduce the
number of elevator by one elevator, while achieving an interval that is just above the
target interval (0.5 seconds in the example above).
It is also worth noting that the speed could be increased to the next preferred
value of 1.6 m/s. This is not unreasonable, and will reduce the time to travel between
terminal landings down to 23.6 s. It results in a solution comprising four elevators
running at a speed of 1.6 m/s and a car capacity of 630 kg/8 persons. It provides a
handling capacity of 12%, an actual interval 22.3 s and a car loading of 67% (P =
5.35 passengers).
The example above has illustrated the range of options available to the
designer, and how the use of personal judgement and trial and error can be used to
arrive at a feasible solution.■

5.0 RULE BASED ELEVATOR TRAFFIC DESIGN

In cases where the design is wasteful or non-compliant (examples were given in
Examples 4 and 5 respectively) it is possible to use rules in order to improve the
design. Rule based design methods have been presented for elevator traffic design
([6], [7]) and for software configuration ([11]).
Two examples are presented in this section to show how the use of rules can
be used to improve wasteful design and address the problem of non-compliant
designs.

Coarse and fine tuning of the results

1) Once an interval is obtained from the software, we can then proceed to amend
the result, as follows:

a) If the actual interval is much higher (e.g., 50 s) than the target interval, then
increase the number of elevators.
b) If the actual interval is much lower (e.g., 10 s) than the target interval, then
decrease the number of elevators.
c) If the actual interval is slightly higher (e.g., 35 s) than the target interval,
then increase speed of the elevators.
d) If the actual interval is slightly lower (e.g., 25 s) than the target interval,
then increase speed of the elevators.

2) After the last step is done, then (and only then) we examine the value of the car
loading, and amend the car capacity as follows:

a) If the car loading is too low (e.g., 20%) then reduce the car capacity.
b) If the car loading is just below 80% then do nothing.
c) If the car loading is above 80%, then increase the car capacity.

Example 7: Improving a Wasteful Design

Revisit Example 4 and address the problem of excessive handling capacity.

In order to address the problem of excessive handling capacity, the following rule will
be used:

If
the handling capacity (HC%) is much larger than the expected arrival rate (AR%)
and
the actual interval is smaller than the target interval
then
reduce the number of passengers, P

The value of P used in the calculations was 20.8. According to the rule above, the
value of P should be reduced. It shall be reduced to 12 passengers. The new value
of P shall be used in order to reduce the value of the handling capacity. The new
value of H, S and the round trip time are evaluated below.

9   i 12 
H = 10 − ∑     = 9.633 (43)

i =1   10  


  1  
12

S = 10 ⋅ 1 − 1 −  = 7.176 (44)
  10  
 

+ (7.176 + 1) ⋅ (2.6 + 2 + 3) + 2 ⋅12 ⋅1.2

4.3
RTT = 2 ⋅ 9.633 ⋅
1.6 (45)
= 142.715 s

Dividing this new value of the round trip time by the target interval and rounding up,
gives the required number of elevators:

 RTT  142.715 
L= =  = 5 (46)
 INTtar   30

The new value of P has lead to a reduction of the number of elevators by one
elevator (from six down to five elevators). However, it is necessary to check that the
handling capacity is more than the arrival rate.
The new actual interval will be:

RTT 142.715
INTact = = = 28.543 s < 30 s = INTtar (47)
L 5

And the new handling capacity of the resulting system will be:

300 ⋅ P ⋅ L 300 ⋅12 ⋅ 5

HC % = = = 12.6% > AR% = 12% (48)
RTT ⋅ U 142.715 ⋅1000

Thus by reducing the number of passengers, P, one elevator has been saved and
the design is still compliant.

Example 8: Correcting a Non-Compliant Design

Revisit Example 5 and change the elevator traffic system to make it compliant.

In this case, the following rule will be used:

If
the handling capacity (HC%) is smaller than the expected arrival rate (AR%)
and
the actual interval is larger than the target interval
then
increase the number of elevators

Still using a number of passengers, P, equal to 8, it is necessary to find the required

number of elevators as follows:

 RTT  101.781
L= = =4 (49)
 INTtar   30 

The new actual interval and the new handling capacity are:

RTT 101.781
INTact = = = 25.44 s < 30 s = INTtar (50)
L 4

300 ⋅ P ⋅ L 300 ⋅ 8 ⋅ 4
HC % = = = 16.84% >> AR% = 13% (51)
RTT ⋅ U 101.781 ⋅ 560

Although this change in the number of elevators has resulted in a compliant design, it
is considered wasteful as the handling capacity is much larger than the arrival. Thus
the rule presented in Example 7 can be used. An attempt will be made to reduce the
number of passengers, P, down to 6.2. The value of H, S and the round trip time are
re-evaluated based on the new value of P.

6   i  6.2 
H = 7 − ∑     = 6.46 (52)

i =1   7 



  1  6.2 
S = 7 ⋅ 1 − 1 −   = 4.3083 (53)
  7 
 

+ (4.3 + 1) ⋅ (2.6 + 2 + 3) + 2 ⋅ 6.2 ⋅1.2

4.5
RTT = 2 ⋅ 6.46 ⋅
1.6 (54)
= 91.56 s

Re-checking the required number of elevators, shows that four elevators are still
necessary. The new interval and the new handling capacity are re-evaluated below.

RTT 91.56
INTact = = = 22.89 s < 30 s = INTtar (55)
L 4

300 ⋅ P ⋅ L 300 ⋅ 6.2 ⋅ 4

HC % = = = 14.5% > AR% = 13% (56)
RTT ⋅ U 91.56 ⋅ 560
Thus both criterion of quality and quantity of service have been achieved, and the
design is now compliant. It is worth noting that two rules have been used in this
example.

6.0 OTHER RULES

Other rules of thumb are presented in this section.

The rule of thumb on speed selection is to divide the total travel by 20. This is
assuming that the elevator can travel between the main terminal and the top most-
floor in 20 seconds (approximately). Once the two have been divided we need to find
the nearest preferred speed (e.g., 1.6 m/s, 2.5 m/s..). A more detailed discussion on
speed selection can be found in the article on elevator kinematics [3].

There are also rules of thumb on the maximum number of elevators in a group and
arrangements. The maximum number of elevators in one group cannot exceed eight
elevators for conventional control systems and 12 for destination control systems. It
is always good practice to keep elevators as near as possible to each other and
group them, in order to achieve the best passenger service and the lowest interval.

7.0 FINDING THE NUMBER OF ELEVATORS TO MEET THE ARRIVAL RATE

In all of the previous examples, we found the required number of elevators in order to
meet the target interval and then carried out a check on the arrival rate. In this
section, the number of required elevators is found in order to first meet the arrival
rate and then the actual interval is checked. There is no guarantee that the target
interval will be met.

Example 9: Finding the number of elevators to meet the arrival rate first
The parameters for an office building are shown below.

a. Passenger arrival rate is 15%.

b. Target interval 25 s (high standard offices).
c. Number of floors above ground is 14 floors.
d. Car capacity is 21 persons (1600 kg).
e. Floor height is 4.5 m (finished floor level to finished floor level).
f. Rated speed, v, is 1.6 m·s-1.
g. Rated acceleration, a, is 1 m·s-2.
h. Rated jerk, j, is 1 m·s-3.
i. Passenger transfer time out of the car is 1.2 s.
j. Passenger transfer time into the car is 1.2 s.
k. Door opening time is 2 s.
l. Door closing time is 3 s.
m. Advanced door opening is 0.5 s.
n. Start delay is 1 s.
o. Total building population of 1100 persons.
p. Equal floor populations.

Find the number of elevators required to achieve the required arrival rate. Then find
the actual interval.
Solution
The equation for the round trip time ( τ ) can be written as follows:

df   
 + (S + 1) ⋅  t f − f + t do + t dc + t sd − t ao  + P (t pi + t po )
d
RTT = 2 ⋅ H ⋅  (57)
 v   v 

Where:

We need to check that the elevator will attain top speed in a one floor journey ([3],
[16]). We use the following equation as a check:

 a 2 v + v 2 j   1 ⋅ 1.6 + 1.6 2 ⋅ 1 
d = 4.5 ≥   =   = 4.16 (58)
 aj   1 

So the top speed of 1.6 m/s will be attained in a one floor jump. In this case, the time
taken to complete a one floor journey, tf will be:

d v a 4.5 1.6 1
tf = + + = + + = 5.4 s (59)
v a j 1.6 1 1

We first start by finding the value of P based on 80% car loading. P works out as
16.8 passengers.
Moving on to calculate H and S as follows:

  1    1  
P 16.8

S = N ⋅  1 −  1 −   = 14 ⋅  1 −  1 −   = 10 (60)
  N     14  
  

N −1 P 16.8
 i  13
 i 
H = N − ∑   = 14 − ∑   = 14 − 0.288 − 0.075 − 0.0174...... = 13.62 (61)
i =1  N  i =1  14 

Substituting in the RTT equation gives:

df   
+ t do + t dc + t sd − t ao  + P (t pi + t po )
df
RTT = 2 ⋅ H ⋅   + (S + 1) ⋅  t f −
 v   v 
 4.5    (62)
 + (10 + 1) ⋅  5.4 − + 2 + 3 + 1 − 0.5  + 16.8 ⋅ (1.2 + 1.2 )
4.5
= 2 ⋅ 13.6 ⋅ 
 1.6   1.6 
= 76.5 + 89 + 40.3 = 205.82 s

In order to meet the expected arrival rate of 15%, the number of elevators required
can be calculated as follows:

 AR% ⋅ U ⋅ RTT   0.15 ⋅ 1100 ⋅ 205.82 

L=
 300 ⋅ P  =  300 ⋅ 16.8  = 6.738 = 7 (63)
Thus seven elevators are required in order to achieve the handling capacity that
meets the expected arrival rate. In practice the handling capacity achieved will be:

300 ⋅ L ⋅ P 300 ⋅ 7 ⋅ 16.8

HC % = = = 15.6% (64)
RTT ⋅ U 205.82 ⋅ 1100

…which is more than the expected arrival rate of 15%.

The achieved interval will be:

RTT 205.82
INTact = = = 29.4 s (65)
L 7

…which is more than the 25 s target interval stipulated by the user. This example
has shown that it is possible to select the required number of elevators to meet the
quantity of service, but this does not guarantee that the quality of service will be
achieved.

The Case of Multiple Entrances

All the examples in this paper have been based on a single entrance. It is possible to
deal with the case of multiple entrances as shown in reference [14].

PROBLEMS
Problems 1 to 4 (the following information applies to problems 1 to 4)
For each of the following four cases, design a suitable elevator traffic system, by
finding the suitable number of elevators and their speed, as well as the actual car
loading (as a percentage).
Try to find the most economical solution by looking for the minimum number of
elevators that meets the requirements. Assume equal floor population and equal
floor heights. In using the round trip time equation ignore the fact that the top speed
has not been attained. Select the speed v by dividing the total travel distance and
then rounding down to the nearest preferred speed (1, 1.6, 2, 2.5, 3.15, 4, 5, 6.3, 8,
10 m·s-1).

Find the following:

L the number of elevators
v the rated speed
Pact, the actual number of passengers
INTact the actual interval
HC% the handling capacity

Also calculate the actual car loading. The actual car loading = (Pact/CC)%

Assume that the door opening time is 2 s

Assume that the door closing time is 3 s
Assume that the start delay is 1 s
Assume that the advance door opening is 0.5 s.
Assume that the passenger transfer time is 1.2 s
Assume that the acceleration, a, = 1 m·s-2
Assume that the jerk, j, = 1.5 m·s-3

1. Office Building: An office building has the following parameters.

N= 20 floors above the main terminal.
AR%= 12%
Target interval, INTtar of 30 s
Car capacity, CC, of 26 persons
Floor height, df, = 4.5 m
Total population, U is 1600 persons

Assume one single main entrance.

2. Office Building: An office building has the following parameters.

N= 25 floors above the main terminal.
AR%= 10%
Target interval, INTtar of 25 s
Car capacity, CC, of 26 persons
Floor height, df, = 4.5 m
Total population, U is 1000 persons

Solve for the following two cases:

a) Single entrance without any basements.
b) One main entrance and three car park basements, each basement with an arrival
percentage of 6% for each.

3. Office Building: An office building has the following parameters.

N= 12 floors above the main terminal.
AR%= 15%
Target interval, INTtar of 30 s
Car capacity, CC, of 26 persons
Floor height, df, = 4.5 m
Total population, U is 800 persons

Assume one single main entrance.

4. Residential Building: A residential building has the following parameters.

N= 30 floors above the main entrance.
AR%= 7%
Target interval, INTtar of 50 s
Car capacity, CC, of 16 persons
Floor height, df, = 3.2 m
Total population, U is 240 persons

Assume one single main entrance.■

Problems 4 to 7
Design suitable elevator traffic systems for the three cases shown below. In each
case, ensure that the design satisfies both the quantity and the quality requirements.
 Problem 4: Problem 5: Problem 6:
N = 6; N = 10; N = 20;
INTtar = 80 s; INTtar = 40 s; INTtar = 30 s;
df = 3.5 m; df = 4.2 m; df = 4.5 m;
v = 1 m.s-1; v = 1.6 m.s-1; v =1.6 m.s-1;
a = 1 m.s-2; a = 1 m.s-2; a = 1 m.s-2;
j = 1 m.s-3; j = 1 m.s-3; j = 1 m.s-3;
tdo = 2s; tdo = 2s; tdo = 2s;
tdc= 3s; tdc = 3s; tdc= 3s;
tpi = tpo 1.2s; tpi = tpo = 1.2s; tpi = tpo 1.2s;
U = 100 person; U = 600 person; U = 1200 person;
AR% = 10%; AR% = 15%; AR% = 12%;

N.B. Always check that the top speed is attained in one floor journey.

REFERENCES
[1] Al-Sharif L. Introduction and Assessing Demand in Elevator Traffic Systems
(METE I). Lift Report 2014; 40(4): 16-24. July/August 2014.
[2] Al-Sharif L. Calculating the Elevator Round Trip Time for the Most Basic of
Cases (METE II). Lift Report 2014; 40(5): 18-30. Sep/Oct 2014.
[3] Al-Sharif L. Intermediate Elevator Kinematics and Preferred Numbers (METE
III). Lift Report 2014; 40(6): 20-31. Nov/Dec 2014.
[4] Al-Sharif L. Four Special Cases in the Calculation of the Elevator Round Trip
Time (METE IV). Lift Report 2015; 41(1): 22-36. Jan/Feb 2015.
[5] Al-Sharif L. Calculating the Elevator Round Trip Time for the General Case
under Incoming Traffic Conditions (METE V). Lift Report 2015; 41(2): 24-37.
Mar/Apr 2015.
[6] N. A. Alexandris, 1988, “LIFTES – An expert system for lift system design”, in
Elevator Technology 2, proceedings of Elevcon ’98, Karlsruhe, Germany, June
1988, pp 1-9.
[7] A. F. Alani, P. Mehta, J. Stonham, R. Prowse, 1995 “Performance optimisation
of knowledge-based elevator group supervisory control system”, in Elevator
Technology 6, proceedings of Elevcon ’95, Hong Kong, March 1995, pp 114-
121.
[8] Al-Sharif L, Abu Alqumsan A M and Abdel Aal O F. Automated optimal design
methodology of elevator systems using rules and graphical methods (the
HARint plane). Building Services Engineering Research and Technology 2013;
34(3): 275-293. August 2013.
[9] Al-Sharif L, Abdel Aal O F, Abu Alqumsan A M and Abuzayyad M A. The
HARint Space: A Methodology for Compliant Elevator Traffic Designs. Building
Services Engineering Research & Technology, 2015; 36(1): 34–50.
[10] Al-Sharif L and Abu Alqumsan A M. Stepwise derivation and verification of a
universal elevator round trip time formula for general traffic conditions. Building
Services Engineering Research and Technology 2015; 36(3): 311-330.
[11] S. R. Sanjeev, 1991, “Knowledge based system to generate lift software”, in the
proceedings of Interlift ’91 Convention, Munich, 23-24 September 1991, pp 90-
98.
[12] Barney G C and Al-Sharif L. Elevator Traffic Handbook: Theory and Practice.
2nd Edition. Routledge. 2016.
[13] CIBSE, “CIBSE Guide D, “Transportation Systems in Buildings”, Third edition,
The Chartered Institute of Building Services Engineers, 2005.
[14] Al-Sharif L. The effect of multiple entrances on the elevator round trip time
under up-peak traffic. Math Comput Model 2010; 52(3–4): 545–555.
[15] Al-Sharif L, Dahyat H and Al-Kurdi L. The use of Monte Carlo Simulation in the
calculation of the elevator round trip time under up-peak conditions. Building
Services Engineering Research and Technology 2012; 33(3): 319–338.
[16] Peters RD. Ideal lift kinematics: derivation of formulae for the equations of
motion of a lift. Int J Elevator Eng 1996; 1: 60–71.
[17] Al-Sharif L, Abu Alqumsan A M and Khaleel R. Derivation of a Universal
Elevator Round Trip Time Formula under Incoming Traffic with Stepwise
Verification. Building Services Engineering Research and Technology 2014;
35(2): 198–213.