Available online at www.sciencedirect.

com

Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579

EWGT 2012

15th meeting of the EURO Working Group on Transportation

Planning of an integrated urban transportation system based on

macro – simulation and MCDM/A methods
Szymon FIEREK*, Jacek ZAK
Poznan Univercity of Technology, Logistics Division, 3 Piotrowo street, Poznan 60-965, Poland

Abstract

The paper presents the planning process of an integrated urban transportation system (IUTS) for a medium sized metropolitan
area. The proposed approach has a universal character and can be applied by city planners, traffic engineers and municipal
authorities in strategic planning of urban transportation systems and designing advanced transportation solutions. It focuses on
the integration of both public and private transportation within the metropolitan area. The overall objective of the integration
is to enhance the traveling standards within the metropolitan area resulting in the change of the residents’ traveling habits and
behavior and modification of the modal split. The proposed planning process is based on macro simulation and Multiple
Criteria Decision Making (MCDM/A) methodology. The macro simulation software VISUM is applied to generate different
scenarios (variants) of an IUTS. These variants are designed heuristically, based on the analysis of the existing urban
transportation system and evaluated by a consistent family of criteria which takes into account interests of different
stakeholders. In the evaluation process the Electre III/IV method is applied. Based on the questioner survey and interviews
the preferences of the city authorities - decision maker (DM) are recognized and transferred into the preference model
characteristic for the applied method. Computational experiments lead to the final ranking and selection of the most desirable
IUTS.

© 2012
© 2012 Published
The Authors. Published
by Elsevier Ltd.bySelection
Elsevier and/or
Ltd. Selection and/or
peer-review peer-review
under under
responsibility ofresponsibility of the Program Committee.
the Program Committee
Open access under CC BY-NC-ND license.
Keywords: Transport integration; urban transportation system; simulation; MCDM/A.

1. Introduction

The constant grow of wealth of the society in the urban areas results in a natural tendency to move residents’
dwellings away from city centres to the suburbs. This effect is called “urban sprawl” or suburbanization (Jackson,
1985), and it has spread around the world for many years. It is currently under way in many Eastern Europe cities
which develop dynamically since the early 90s of last century due to political and economical transformation in

* Corresponding author. Tel.: +48 61 6652716; fax: +48 61 665 2230.

E-mail address: [email protected]

1877-0428 © 2012 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of the Program Committee
Open access under CC BY-NC-ND license. doi:10.1016/j.sbspro.2012.09.774
568 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579

the region.
The “urban sprawl” is clearly linked with transferring certain activities from the city centre to its periphery and
it is accompanied by a number of economical, social and environmental consequences. One of the effects of
suburbanization is the development of road infrastructure and the formation of the so-called “city bedroom”,
which needs a reasonable connection with other parts of the city.
In the majority of cases suburbs are poorly-connected with city centres by public transportation. Thus, the
residents of suburban areas travel on a regular basis to the city centre (schools, jobs, places of entertainment) by
private cars. This results in the overall increase of movements and congestion of the roads connecting suburbs
with the city centres, especially during peak hour. This situation is very burdensome for various stakeholders,
such as: road users, residents of urban areas and public transportation operators. High level of congestion,
generated in particular in the areas of heavy traffic, in addition to producing high travel cost, leads to increased
emissions of noise and pollution. These effects contribute to the progressive degradation of the environment and
urban values (Millsa, 1984). Therefore, transportation between suburban and urban areas plays a key role in the
development of the whole metropolitan area.
One of the ways to reduce high level of congestion and its negative effects is promotion of public
transportation which can take over a substantial portion of travels previously carried out by private cars. This task
is very challenging due to the fact that the private transportation has a significant advantage over public
transportation i.e. possibility of the "door to door" trips. Thus, it is very important to increase the competitiveness
of public transportation to achieve the above mentioned goals. One of the ways to do it is to integrate public and
private transportation within the urban areas. This integration can be achieved at relatively low costs due to the
fact that it is focused on linking a well-developed road infrastructure within suburban areas and dense public
transportation system in the city. Therefore the integration of public and private transportation is one of the key
issues promoted by EU (European Commission, 2004) as a tool for improvement of urban transportation systems.
Many European Union research projects (EU-Spirit, 2001; Guide, 1999; Link, 2007; Mimic, 1999; Pirate,
1999; Propolis, 2004; Prospects, 2003; Transplus, 2003; Quattro, 1997) have been focused on the development of
solutions that would enhance the flexibility of urban transportation systems, including: coordination and
integration between transportation modes, system oriented planning and designing of the transportation network,
introduction of demand responsive, passenger oriented transportation system.
Transport integration is a widely discussed subject (Bonsall, 2000; Hull, 2005; Keuschel et al., 2011; May et
al., 1995; May et al., 2006; May et al., 2012, Mua R., 2012; Parkhurst et al., 2002; Potter et al., 2000; Preston,
2010; Walton et al., 2003; Zhang et al., 2007). The notion of the transport integration may be perceived in
different way. Preston and others (2008) define urban transport integration as follows: “The organizational
process through which the planning and delivery of elements of the transport system are brought together, across
modes, sectors, operators and institutions, with the aim of increasing economic and social benefits”. May and
others (1995, 2006) divide transport integration into four types: integration between different modes; integration
between policy instruments involving infrastructure provision, management, information and pricing; integration
between transport measures and land use planning measures; and integration with other policy areas such as
health and education.
The most commonly cited objectives of transport integration are (May, 1991; Preston, 2010): efficiency in the
use of resources; improved accessibility; environmental protection; increased safety.
To reach these objectives different transportation solutions should be considered, including: Park & Ride, Park
& Go and Bike & Ride parking lots, construction of passengers’ friendly transferring nodes, utilization of shared
transportation infrastructure by different transportation modes (e.g. trams and trains used on the same tracks, High
Occupancy private Vehicles – HOVs allowed to use bus lanes), application of advanced electronic (web-based
and mobile) passengers travel planners.
The design of the most rational configuration of transportation solutions integrating urban transportation
systems can be supported by different tools and methods including simulation. Simulation applied in the
 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579 569

transportation environment assists traffic engineers and DMs in forecasting traffic flows and utilization of
transportation infrastructure in a certain transportation system. It helps to generate a set of evaluation measures
that asses the transportation system behavior and operations. These parameters can be applied in the evaluation
process of different alternative transportation solutions.
The most commonly used characteristics, applied for evaluation of urban transportation systems, including
integrating solutions are as follows (ĩak, 2005): riding time, waiting time, operating frequency or headway,
transferring frequency, safety, comfort, accessibility, noise and vibration, comfort of aboarding and alighting. In
many specific conditions these criteria should be adjusted to the local circumstances and should take into
consideration interests and expectations of different stakeholders.
N. Caliskan (2006) distinguishes two main groups of stakeholders in the urban transportation system: local
authorities and other users of the system. G. Tzeng and M. Shiau (1988) indicate a significant role played by
passengers and operator of the transportation system. J. Zak (1999) and J. Zak and T. Thiel (2001) specify the
passengers, operator/ operators of the transportation system and local authorities as major entities interested in
efficient operations of the urban transportation system. The interests of these groups are often contradictory.
Thus, it becomes necessary to search for compromise solutions, that would satisfy (at least partially) all the
parties interested in the integration of urban transportation systems. It is, therefore, legitimate using multiple
criteria decision making/ aiding methodology for the evaluation of transportation solutions focused on integration.
In this article, the authors describe the methodology of designing and evaluating the development scenarios of
IUTS based on macro-simulation and multiple criteria decision making/aiding (MCDM/A). The authors consider
one type of the above mentioned integrations, which is integration between different transportation modes, and
specifically between public and private transportation. In their understanding this integration includes all
activities enabling easy transfer between given transportation modes (e.g. cars, bikes, motorbikes, buses, trams,
metro, etc.) and smooth door – to – door passengers’ travels within metropolitan areas.
The paper is composed of six sections. The introduction presents the literature survey and the background of
the topic considered. In section 2 the research methodology is presented, while section 3 describes the constructed
variants. In section 4 the evaluation of variants is carried out and the results of computational experiments are
presented. Final conclusions are drawn in section 5. The paper is completed by a list of references.

2. Research methodology

This section of the paper presents the methodological background of the research. Two principal theories
supporting the planning process of IUTS i.e. traffic simulation (subsection 2.1) and MCDM/A (subsection 2.2)
are described.

2.1. Traffic simulation

Simulation is a popular and effective operations research technique used to analyze a wide range of
dynamically changing systems (Sokolowski and Banks, 2009). Researchers define the simulation as a controlled
statistical sampling technique (Hiller, Lieberman, 2001). The essence of simulation is to carry out a series of
computational experiments, using various input data introduced into a simulation model, that describes the
operations of the real system and to generate a set of output data that characterizes and assess (the considered
system) it. Simulation allows us to check, monitor and evaluate the behavior of the real system under different
realistic conditions, in an artificial computer – based lab environment.
Based on the Highway Capacity Manual (2000), traffic simulation is “a computer program that uses
mathematical models to conduct experiments with traffic events on a transportation facility or system over
extended periods of time”. It can describe the entire transportation system (e.g. urban, regional or national) or its
selected part or component, such as: several intersections, a motorway segment and/or a roundabout. The
570 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579

simulation model consists of two mutually interrelated components, i.e.: information on demand that
characterizes the passengers’ travel needs and information on supply that describes transportation network, traffic
zones (e.g. residential areas) and public transportation vehicles, etc. Traffic simulation is applicable in
(Liebermann & Rathi, 1997):

• designing and evaluating alternative transportation solutions,

• training personnel (e.g. operators in traffic management centers),
• safety analysis (e.g. reconstructing a road accident).

Depending on their accuracy and scope the simulation models can be divided into (Liebermann & Rathi,
1997):

• Micro-simulation models that describe traffic at high level of detail and distinguish single, separate units in the
traffic flow (different types of vehicles, pedestrians) and mutual interactions between them. They are usually
applied for the detailed analysis of limited segments of transportations systems. The most popular micro-
simulation tools are Vissim, Aimsun, Corsim.
• Meso-simulation models that describe traffic at an intermediate level of detail and distinguished particular
units in the traffic flow but do not take into account interactions between them. They are less precise and
usually applied for the components of transportation systems covering larger areas. The most popular meso-
simulation tools are Netflo, Dynasmart, Transims.
• Macro-simulation models that describe traffic at a high level of aggregation, as a uniform traffic flow. They
are based on deterministic relationships between the quantities characterizing the traffic flow such as: volume,
speed and density. Macroscopic simulation has been developed to model an entire transportation network
and/or system. The most popular programs that simulate the movement in the macro scale are Visum, Transyt,
Emme, Transcad.

Taking into account the above-mentioned, different types of simulation models, it is important to choose the
most suitable tool for the simulation of the considered transportation variants/ solutions applied in a specific
transportation system. The selection of the computer-based simulation software depends on many factors, such
as: the nature and the territorial scope of the analysis, the possibility and accuracy of representing different
objects encountered in the real system, the ability of modeling the behavior of different components of the
system, the availability of the required information/ data (e.g. traffic volume, road capacity, environmental
measures, etc.). Considering the above-mentioned aspects, the authors of this article used macro-simulation tool –
Visum to design and evaluate the development scenarios of IUTS.

2.2. The Methodology of Multiple Criteria Decision Making/Aiding (MCDM/A)

Multiple criteria decision making/ aiding is a dynamically developing field which aims at giving the decision-
maker (DM) some tools in order to enable him/her to advance in solving a complex decision problem, where
several – often contradictory – points of view must be taken into account (Vincke, 1992). In contrast to the
classical techniques of operations research, multicriteria methods do not yield “objectively best” solutions,
because it is impossible to generate such solutions which are the best simultaneously, from all points of view.
The methodology of MCDM/A is a set of rules that are applied in the process of solving the so called multiple
objective decision problems, i.e. situations in which, having defined a set A of actions and a consistent family of
criteria F one wishes to:

• Determine a subset of actions considered to be the best with respect to F (choice problem).
• Divide A into subsets according to some norms (sorting problem).
 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579 571

• Rank actions of A from best to worst (ranking problem).

Based on the research carried out by Zak (2004, 2005) choice and ranking problems constitute the most
important categories of the transportation decision problems.
The MCDM/A methodology clearly identifies the major participants of the decision making/ aiding process,
such as: the decision maker (DM), the analyst and the interveners (stakeholders). The process of solving a
multiple objective decision problem is based on the application of computerized tools and methods. Those
methods are usually classified as follows (Figueira et. al., 2005):

• methods of the American inspiration, based on the utility function e.g. AHP, UTA,
• methods originated in Europe, based on the outranking relation (e.g. Electre III/IV methods, Promethee I and
II , Oreste,
• interactive methods e.g.: GDF, SWT, Steuer, STEM, VIG.

In this paper the Electre III/IV method is applied to evaluate alternative transportation solutions (variants) for
the IUTS.
Electre III/IV method belongs to a family of Electre methods, proposed by Roy (1990b) and it is based on a
binary outranking relation (Skalka, 1986; Vincke, 1992). In this method, the basic set of data is composed of the
following elements: a finite set of variants A, a family of criteria F, and the preferential information submitted by
the DM. The preferential information is defined in the form of criteria weights - w and the indifference - q,
preference - p and veto - v thresholds (Skalka, 1986). The thresholds define the following intervals of preference
between variants on each criterion: indifference (up to q), weak preference (between q and p), (strong) preference
(between p and v) and incomparability (beyond v). Variants a and b are considered indifferent if the difference
between their evaluations f(a) and f(b) on a specific criterion is so small (smaller than q) that the DM cannot
make any distinction between variants. Variant a is weakly preferred against variant b if the difference between
their evaluations f(a) and f(b) on a specific criterion is noticeable to the DM (between q and p) but he/she is
hesitant to prefer one of them. Variant a is strongly preferred against variant b if the difference between their
evaluations f(a) and f(b) on a specific criterion is substantial to the DM (between p and v) and he/she is convinced
that a is preferred to b. Variants a and b are incomparable if the difference between their evaluations f(a) and f(b)
on a specific criterion is so large (larger than v) that the DM cannot consider them as comparable objects.
The outranking relation in the Electre III/IV method is built on the basis of the so called concordance and
discordance tests. In the concordance test a concordance matrix, composed of the global concordance indicators
C(a, b), is constructed. In the discordance test, a discordance index Dj(a,b) for each criterion j is calculated. The
outranking relation indicates the extent to which ”a outranks b” overall. This relation is expressed by the degree
of credibility d(a,b), which is equivalent to the global concordance indicator C(a,b) weakened by the discordance
indexes Dj(a,b). The values of d(a,b) are from the interval [0,1]. Credibility d(a,b) = 1 if and only if the assertion
aSb (“a outranks b”) is well founded, d(a,b)=0 if there is no argument in favor of aSb (not aSb – “a does not
outranks b”). The definition of d(a,b) results in the construction of the credibility matrix based on which the
method establishes two preliminary rankings (complete preorders) using a classification algorithm (distillation
procedure). During this procedure one can obtain a descending and an ascending preorder. In the descending
distillation the ranking process starts from the selection of the best variant, which is placed at the top of the
ranking. In the ascending distillation the variants are ranked in the inverse order. The final ranking is generated as
an intersection of the above mentioned complete preorders. It can be presented either in the form of the ranking
matrix or in the form of the outranking graph. The following situations can be distinguished there: indifference
(I), preference (P), lack of preference (P-) and incomparability (R).
572 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579

3. Construction of variants

As described in section 2 the macro–simulation software Visum has been utilized to generate alternative
solutions for the IUTS. The following four phases (four step approach – (McNally, 2000). have been carried out
to model all the variants of integration of the urban transportation system within the metropolitan area, including:

• Trip generation, focused on constructing the traffic zones and recognizing their travel potential, and defining
origins of all passengers’ trips in each traffic zone† .
• Trip distribution featured by determination of all passengers’ trips between any pair of zones
• i (origins - O) and j (destinations - D). The result of this step is a construction of a complete O-D matrix
(McNally, 2000).
• Mode choice concentrated on the allocation of previously defined trips to available transportation modes.
• Route assignment consisting in assigning the distribution of O-D matrix to the transportation network, which
results in generating the volume of passengers’ and vehicles’ flows on particular links of the network.

Taking into account the overall objective of the transformation of the urban transportation system, which is its
integration (in particular integration between private and public transportation), the authors proposed a number
infrastructural and organizational changes to enhance coherency between transportation modes and overall
standard of passengers’ travels within urban area. The proposed solutions such as: Park&Ride parking lots,
Bike&Ride parking lots, shared infrastructure for different transportation modes, combination of road and inland
water transportation are describe in table 1. All constructed variants, presented in table 1, have been designed
based on the analysis of the existing transportation system.
The considered urban transportation system covers an area of 300 sq. km inhabited by a population of 550 000
people. It is based on three internal bypasses called “rings” as well as on external bypass connected with
highway. Total length of the roads within the urban area amounts to 1150 km. Mass transit system in the urban
area based on a tram network (20 daily routes and one night route) supplemented by a dense bus network (80
daily routes and 20 night routes) is relatively well developed. The major disadvantage of the analysed
transportation system is poor integration between public and private transportation.

Table 1. Transportation solutions proposed in particular variants that support integration of the urban transportation system

Variant 1 (V1) – Park and Ride oriented IUTS

This variant assumes the construction of:
• Three parking lots for private cars (Park and Ride – P&R system) in locations A, B and C shown on fig 1.
Variant 2 (V2) - Road oriented IUTS
This variant includes the construction of:
• Five P&R parking lots (locations A – E on fig 1.).
• Three Park & Go (P&G) parking lots (locations 1 – 3 on fig 1.).
• 14 Bike & Ride (B&R) parking lots.
• Shared lanes for HOVs, buses and taxis, separated from a regular traffic. Two corridors through the whole city – horizontal
(east - west) and vertical (north – south) – marked on fig 1. and denominated as R1 – R5.
• Providing access to tram railroads for HOVs, buses and taxis (in all segments where it is technically possible).
Variant 3 (V3) - rail oriented IUTS

† Traffic zone – a limited area featured by a homogenous: land use policy and similar category of infrastructure, inhabited by a number of
people characterized by common travel behavior (McNally, 2000).
 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579 573

This variant provides for the construction of:

• Eight P&R parking lots (locations A – G on fig 1.).
• Three P&G parking lots (locations 1 – 3 on fig 1.).
• 14 B&R parking lots.
• Shared lanes for HOVs, buses and taxis, separated from a regular traffic – marked on fig 1. and denominated as R3 – R5.
• Five routes of low floor, double system (tram-train vehicles, operating on tram infrastructure within urban area and heavy
railway infrastructure beyond the city center, marked on fig 1. and denominated as T1 – T5.
• Extension of two tram routes – marked on fig 1. and denominated as T6 and T7.
• Providing access to tram railroads for HOVs, buses and taxis (in all segments where it is technically possible).
Variant 4 (V4) - waterborne oriented IUTS
This variant is featured by:
• Eight P&R parking lots (locations A – G on fig 1.).
• Four P&G parking lots (locations 1 – 4 on fig. 1.).
• 14 B&R parking lots.
• Shared lanes for HOVs, buses and taxis, separated from a regular traffic – marked on fig 1. And denominated as R3 – R5.
• One route of low floor, double system tram-train vehicles (operating on tram infrastructure within urban area and heavy
railway infrastructure beyond the city center, marked on fig 1. and denominated as T4.
• One water tram route on the river – marked on fig 1. and denominated as W1.
• Providing access to tram railroads for HOVs, buses and taxis (in all segments where it is technically possible).
Variant 5 (V5) – hybrid IUTS
This variant envisages the most radical transformation and provides:
• Nine P&R parking lots (locations A – H on fig. 1.).
• Four P&G parking lots (locations 1 – 4 on fig. 1.).
• 14 B&R parking lots.
• Three lines of low floor tram-train – trams using both tram infrastructure within urban area and heavy railway infrastructure
beyond the city – marked on fig 1. and denominated as T1, T2 and T4.
• Extension of one tramway route – marked on fig 1. and denominated as T6.
• One water tram route on the river – marked on fig 1. and denominated as W1.
• Providing access to tram railroads for HOVs, buses and taxis (in all segments where it is technically possible).

The existing urban transportation system (variant V0) with all the above mentioned changes characteristic for
variants V1, V2, V3, V4 and V5 are presented in fig 1.

4. Evaluation of variants

The above characterized variants of the integrated urban transportation system have been evaluated by a
consistent family of criteria. Based on the literature review presented in section 1, the authors of the paper
considered different parameters and characteristics that could be used to evaluate an IUTS and analyzed their
applicability in a considered transportation system. As a result they proposed the below mentioned measures as
evaluation criteria. Four of them (C1, C4, C5, C6) have a universal character and are usually applied as generic
characteristics of the evaluation of transportation projects. Two remaining criteria (C2, C3) are specifically
designed to evaluate the integration the urban transportation system. The proposed criteria are as follows:

• Travel time – C1 [minutes] – is defined as time required to carry out within the metropolitan area, a
passenger’s “door to door” travel of an average distance. It is assumed that in the passenger’s travel both
private and public transportation means are used. This minimized criterion is important for passengers, drivers
and other road users (bikers, pedestrians).
574 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579

Fig. 1. The scheme of analyzed IUTS with proposed transportation solutions supporting modal integration (in particular integration between
public and private transportation).

• Transferring time – C2 [minutes] – is defined as an average time spend by a passenger for all transfers during
his/her “door to door” travel within the metropolitan area. This minimized criterion, important for passengers
measures their convenience during travels in particularly during transfers.
• Effectiveness of integration – C3 [%] – is defined as a percentage share of travels carried out by both private
and public transportation means in a total number of metropolitan travels within an urban area. This
maximized criterion is particularly important for municipal authorities and an operator of the public
transportation system.
• Investment profitability – C4 [%] – is defined as an internal rate of return (IRR) or in other words the interest
rate at which the discounted investment costs and incomes generated by the urban transportation system
balance in a certain time horizon (20 years). This maximized criterion is important for municipal authorities.
 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579 575

• Environmental friendliness – C5 [pts] – is defined as a measure that characterize the level of environmental
friendliness of the urban transportation system; this criterion is expressed in points assign by experts for
different solutions that eliminate threats for the environment (pollution, noise) and promote environmentally
friendly - “clean & green” transportation means. This maximized criterion is important for the residents of the
metropolitan area.
• Safety – C6 [pts] – is defined as an inverse of the risk (probability) of accidents correlated with the condition
of transportation infrastructure, proposed infrastructural solutions and traffic volume. It is is measured in
points, assigned by experts for an overall safety of urban transportation system. This maximized criterion is
important for all stakeholders.

All variants (V0 – V5) have been evaluated by this family of criteria. In the majority of cases the evaluations
have been generated by the traffic simulator Visum. For selected criteria the expert evaluation has been carried
out. The evaluation matrix is presented below (table 2).

4.1. Modelling of the DM’s preferences

Based on the survey and interviews with the DM, his/her preferences have been recognized and transformed
into the model of preference characteristic for the Electre III/ IV method. As described in section 2.2 the
Electre III/IV method utilizes a preference model based on weights of criteria – w and thresholds q, p and v.
Those thresholds represent the sensitivity of the DM to the changes of the criteria values. All four values – w, q, p
and v are defined separately for each criterion. The model of the DM’s preferences characteristic for the Electre
III/IV method is presented in table 3.
As one can see (table 3) the most and the least important criteria are: C1 – travel time (weight w=9) and C5 –
environmental friendliness (weight w=4), respectively.

Table 2 Evaluation matrix for different variants of the IUTS in analyzed city

Variants
Criteria
V0 V1 V2 V3 V4 V5
C1 Travel time [min] 45,5 46,4 44,2 40,1 43,1 39,5
C2 Transferring time [min] 12,2 13,1 11,7 11,2 12,3 12,8
C3 Effectiveness of integration [%] 5,1 5,5 5,8 8,6 5,6 9,4
C4 Investment profitability [%] - 4,2 3,1 2,5 3,3 0
C5 Environmental friendliness [pts] 1 2 4 6 5 8
C6 Safety [pts] 3 3 4 6 5 7

Table 3. The model of the DM’s preferences characteristic for the Electre III/IV method.

Criteria q p v w
C1 Travel time [min] 1 3 10 9
C2 Transferring time [min] 0,5 2 6 7
C3 Effectiveness of integration [%] 0,5 2 5 6
C4 Investment profitability [%] 1 3 10 7
C5 Environmental friendliness [pts] 0,5 2 4 4
C6 Safety [pts] 0,5 2 4 6
576 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579

a) b)

Fig. 2. Numerical results of computational experiments carried out by the Electre III/IV method - a) Concordance Matrix, b) Credibility
Matrix

a) b)

Fig. 3 Final results of computational experiments carried out by the Electre III/IV method a) Ranking Matrix b) Final Graph

4.2. Computational experiment

The computational experiment has been carried out with the application of the computer implementation of
the Electre III/IV method. As mentioned in section 2.2 the final ranking has been generated based on the
calculation of the outranking relation. To this end Concordance Indicators C(a,b) are computed, which results in
the generation concordance matrix. The Concordance Indicators define the mutual relationships between variants
(see Fig. 2a). In the analysed case, e.g. C(V2,V3)=0.36 and C(V3,V2)=1 suggest that there is a lower chance for
variant V2 to outrank variant V3 and higher chance associated with the inverse order, i.e. that V3 outranks V2.
After having constructed the concordance matrix the procedure computes discordance indexes on specific criteria
and then constructs the outranking relation. This relation is expressed by the degree of credibility, included in the
credibility matrix (Fig.2b).
In the next step of Electre III/IV method two preliminary rankings (complete preorders) are generated using a
classification algorithm (distillation procedure). The intersection of these preorders constitutes the final ranking.
It can be presented in the form of the ranking matrix – Fig. 3a or in the form of the graph (Fig. 3b).
The best solution (variant V3) is preferred against all the remaining variants (relation P in the ranking matrix)
 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579 577

and placed at the top of the final graph. Variants V2 and V4 are indifferent (relation I in the ranking matrix and
placement in the same box in the final graph) and incomparable with V5 in the same time (relation R in the
ranking matrix, which corresponds to the lack of connection between variants in the final graph).

5. Summary and conclusions

The paper presents the application of macro-simulation and MCDM/A method to the design and evaluation of
the variants – development scenarios of IUTS. Various components – transportation facilities enabling
integration between transportation modes – have been proposed. Simulation models of all variants (including
variant V0 – current transportation system) have been constructed and simulated in the macro-simulation
software – Visum. The application of Visum resulted in the generation of a set of parameters and measures that
have been used in the computation of all previously defined criteria. As a result of a computational experiment
carried out by the Electre III/IV method allowed to generate the final ranking of variants.
From a methodological point of view the proposed approach for planning of an integrated urban transportation
system has a universal character and can be applied in a variety of transportation systems. It is composed of the
following stages:

• analysis of the existing urban transportation system,

• design of integrating solutions and definition of different variants of IUTS,
• modeling and simulating the behavior of the proposed variants,
• definition of the consistent family of criteria and construction of the evaluation matrix,
• definition of DM’s model of preferences,
• computational experiments leading to the final ranking.

The proposed methodology assumes that for measuring the integration of an urban transportation systems the
defined family of criteria should include measures of integration such as criteria C2 – transferring time and C3 –
effectiveness of integration.
From the practical point of view the results of this project can be summarized as follows:

• Variant V3 – rail oriented IUTS – is the winner of the generated ranking. The top position of variant V3 in the
ranking is the effect of very good performance on each criterion, including: C1 - travel time which value has
been reduced more than 10% in comparison with its equivalent for variant V0, C2 – transferring time which
has the best value among all variants or C3 – effectiveness of integration which has increases from 5,1%
(variant V0) to 8,6%. Despite the relatively low value of criterion C4 - investment profitability, the rail
oriented variant (V3) is the most desirable.
• Variant V0 – current urban transportation system – located at the bottom of the ranking is the worst variant in
the analyzed set. This variant is characterized by poor values of all criteria which correspond which serious
weaknesses of the existing urban transportation system.
• As far as integration of the urban transportation system is concerned the winning variant V3 is characterized
by excellent performance. On criteria C2 – transferring time and C3 – effectiveness of integration that
measure an overall integration of the urban transportation system variant V3 reaches the best and the second
best position respectively.

Further research should be carried out two following directions: detailed analysis of the winning variant
supported by micro-simulation and application of different MCDM/A methods in evaluation of variants
578 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579

Acknowledgements

This research has been co-financed by the grant: “Scholarship support for Ph.D. students specializing in
majors strategic fields focused on the development of the Great Poland region”, Sub-measure 8.2.2 Human
Capital Operational Programme, co-financed by European Union under the European Social Fund.

References

Bonsall P. (2000). Legislating for modal shift: background to the UK's new transport act. Transport Policy 7, 179-184.
Caliskan N. (2006). A Decision Support Approach for the Evaluation of Transport Investment Alternatives. European Journal of Operational
Research, 175 (3), 1696-1704.
European Commission (2004). Sectoral Operational Programme Transport, final version of February 6, 2004 adapted by Republic of Poland
and European Commission.
Figueira J., Greco S. & Ehrgott M. (2005). Multiple Criteria Decision Analysis: State Of The Art Surveys. Springer Verlag, Boston,
Dordrecht, London
Haimes, Y. & Hall, W. (1975). Multiobjectives in Water Resources Systems Analysis: The SWT Method. Water Resources Research, 10,
615-623.
Highway Capacity Manual (2000). Transportation Research Board.
Hiller, F. & Lieberman, G. (2001). Introduction to Operational Research. McGraw – Hill, New York.
Hull A. (2005). Integrated transport planning in the UK: From concept to reality. Journal of Transport Geography 13, 318–328.
Jackson K.T. (1985). Crabgrass Frontier: The Suburbanization of the United States. Oxford University Press, Oxford.
Keuchel S. & Richter C. (2011). Applying Integrated Hierarchical Information Integration to Mode Choice Modelling in Public Transport.
Procedia Social and Behavioral Sciences 20, 875–884.
Liebermann E. & Rathi A.K. (1997). Traffic simulation. In: Revised Monograph on Traffic Flow Theory, 268-270.
McNally M.G. (2000). The Four Step Model. In: Handbook of Transport Modelling, ed. David A. Hensher and Kenneth J. Button, 35-52.
May A.D. & Roberts M. (1995). The design of integrated transport strategies. Transport Policy, 2, 97-105.
May A.D., Jopson A.F. & Matthews B. (2003). Research challenges in urban transport policy. Transport Policy 10, 157–164.
May A.D., Kelly C. & Shepherd S. (2006). The principles of integration in urban transport strategies. Transport Policy 13, 319–327.
May A.D., Kelly C. Shepherd S. & Jopson A. (2012). An option generation tool for potential urban transport policy packages. Transport
Policy , 20, 162–173.
Millsa E. & Price R. (1984). Metropolitan suburbanization and central city problems. Journal of Urban Economics, 15, 1–17.
Mua R., Jong M., Yu B. & Yang Z. (2012). The future of the modal split in China’s greenest city: Assessing options for integrating Dalian’s
fragmented public transport system. Policy and Society 31, 51–71.
Parkhurst G. & Richardson J. (2002). Modal integration of bus and car in UK local transport policy: the case for strategic environmental
assessment. Journal of Transport Geography 10, 195–206.
Potter S. & Skinner M.J. (2000). On transport integration: a contribution to better understanding. Futures 32, 275–287.
Preston J. (2010). What’s so funny about peace, love and transport integration? Research in Transportation Economics 29, 329-338.
Preston J., Marshall A. & Tochtermann L. (2008). On the Move: delivering integrated transport in Britain’s cities.
Roy, B. (1990a). Multiple Criteria Decision Aiding. Wydawnictwo Naukowo-Techniczne, Warsaw.
Roy, B. (1990b). The Outranking Approach and the Foundations of Electre Methods. In: Bana e Costa (Ed.): Readings in Multiple Criteria
Decision Aid. Springer-Verlag, Berlin, 155-183.
Saaty, T. (1980). The Analytic Hierarchy Process: Planning, Priority Setting, Resource Allocation, McGraw-Hill, New York.
Skalka, J., Bouyssou, D. & Bernabeu, Y. (1986). Electre III et IV. Aspects methodologiques et guide d’utilisation. LAMSADE, 25, Paris.
Sokolowski J. & Banks C. M.(eds.) (2009). Principles of Modeling and Simulation: A Multidisciplinary Approach. John Wiley & Sons, New
York.
Tzeng G.M. & Shiau M.A. (1988). Multiple Objective Programming for Bus Operation: A Case Study for Taipei City. Transportation
Research, Vol. B., 22/23, 195-206.
Vincke, P. (1992). Multicriteria Decision – Aid, John Wiley & Sons, New York.
Walton W. & Shaw J. (2003). Applying the new appraisal approach to transport policy at the local level in the UK. Journal of Transport
Geography, 11, 1–12.
ĩak J. & Thiel T. (2001). Multiple Criteria Evaluation of the Development Scenarios of the Mass Transit System. In: Park C.-H., Cho J., Oh
J., Hayashi Y. & Viegas J. (Eds.): CD- Selected Proceedings of the 9th World Conference on Transport Research, Seoul.
ĩak J. (1999). The Methodology of Multiple-Criteria Decision Making in the Optimization of an Urban Transportation System: Case Study of
Poznan City in Poland. International Transactions in Operational Research, 6, 571-590.
 Szymon FIEREK and Jacek ZAK / Procedia - Social and Behavioral Sciences 54 (2012) 567 – 579 579

ĩak J. (2004). Identification of the Most Important Road Transportation Decision Problems. Archives of Transport, 16,(2), 89–109.
ĩak J. (2005). Multiple Criteria Decision Aiding in Road Transportation, Poznan University of Technology Publishing House, Poznan.
Zhang G., Li M. & Wang J. (2007). Application of the Advanced Public Transport System in Cities of China and the Prospect of Its Future
Development. Journal Of Transportation Systems Engineering And Information Technology 7,(5), 24-30.