You are on page 1of 11

African Journal of Business Management Vol. 5(2), pp.

505-514, 18 January, 2011
Available online at
DOI: 10.5897/AJBM10.661
ISSN 1993-8233 ©2011 Academic Journals

Full Length Research Paper

Soft system modeling in transportation planning:
Modeling trip flows based on the fuzzy inference
system approach
J.Jassbi1, P.Makvandi1*, M. Ataei2 and Pedro A. C. Sousa3
Department of Industrial Management, Faculty of Management and Economics, Science and Research Branch,
Islamic Azad University, Tehran, Iran.
College of Business Administration, University of Missouri (UM-SL), St. Louis, USA.
Faculdade de Ciencias e Tecnoiogia, Universidade Nova de Lisboa, Portugal.
Accepted 14 October, 2010

Transportation planning is one the most important problems of urban management systems.
Meanwhile, modeling trip flows between metropolitan zones is vital to a successful transportation
planning. Due to importance of the problem, different models have been developed in recent years but
because of the complex nature of the problem that deals with human behavior and existence of
different independent variables that affect number of trips, it is always hard to develop a model with
acceptable forecasting error that is computationally efficient. In this paper a three phases fuzzy
inference system (FIS) proposed to map social and demographic variables to total number of trips
between origin-destination (OD), pairs. Fuzzy rule bases in the model are in fact the exploration of
transportation experts’ subjective patterns.

Key words: Transportation planning, trip forecasting, fuzzy inference system, fuzzy rule base.


Traffic flows and trip distribution resulted from human divided into exact regions and planners trying to forecast
choices that are affected by social and individual trip flow between regions as OD pairs (i.e. for 20 years
variables of the commuters. Due to this fact that human after). Based on this information, a proper transportation
decision making are more consistent with fuzzy logic in infrastructure such as subway, highway, etc. is designed
comparison with crisp mathematics, it seems that fuzzy and executed to serve the transportation demands. A trip
logic could be a logical tool to map such areas. Modeling could be defined as follows:
a trip distribution system with fuzzy inference systems
would enjoy the exploration of subjective pattern of “To move between origin to destination by
decision makers. Transportation planning is one the basic
concern and problems of many developed and also mode on route in an exact period of time.”
developing countries all around the world. It is important
because it deals with cost, time and security. On the Meanwhile, one of the most important parts of forecasting
other hand, it could be important for governments procedure is to predict the trips flow between a given
because it could be related to the satisfaction of citizens. origin-destination pair. Different models have been
Transportation planning activities are commonly based developed to forecast trip flows in recent years. The key
on the forecasting. Usually, the under studied zone is to a successful forecasting is to recognize the existing
patterns correctly. Obviously, models with minor forecast-
ting errors would be more suitable for transportation
planning but, because of complex nature of the transpor-
*Corresponding author. E-mail: Tel: +98 tation planning problems that deals with human behavior,
21 447 32 947. Fax: +98 21 447 32 947. it is always hard to find the optimal solution. The traditional
506 Afr. J. Bus. Manage.

four-step model has been widely used in travel demand 1997). Diverse kinds of NNs have been proposed in the
forecasting. The standard model includes: trip generation, literature.
trip distribution, modal split and route choice sequentially
in a top-down sequential process (Ortuzar and
Willumsen, 2001). The four-step model is based on the Fuzzy logic in transportation
functions that try to approximate the trips between the
metropolitan zones. The approximation migrates and is Fuzzy logic provides an effective means of dealing with
reinforced from each step to the next one and this could problems involving imprecise and vague phenomena.
lead to final significant calculations deviation. Although Fuzzy concepts enable assessors to use linguistic terms
each step of the transportation planning is important for to assess indicators in natural language expressions and
transportation planners but, trip timing and mode choice each linguistic term can be associated with a
decisions of the commuters are the most important. membership function. Furthermore, fuzzy logic has found
These two decisions along with route choice directly significant applications in management sciences.
determine the temporal distribution of demand expe- However, the gradual evolution of the expression of
rienced on any given piece of transportation infrastructure uncertainty using probability theory was challenged, first
in an urban area (Susilo and Kitamura, 2007). The in 1937 by max black, with his studies in vagueness, then
reliability of forecasting results influences the following with the introduction of fuzzy sets by Lotfi Zadeh in 1965.
steps such as trip distribution, mode split, and route Zadeh’s work had a profound influence on the thinking
choice. Therefore, improved trip generation models are about uncertainty because it challenged not only pro-
needed to improve forecasting precision (Golob, 2000). bability theory as the sole representation for uncertainty,
The trip distribution models have been classified in two but the very foundations upon which probability theory
broad categories as aggregate and disaggregate models. was based: classical binary (two valued) logic (Ross,
The disaggregate models try to explain individuals’ 2004). Without denying the importance of binary logic as
behaviors in selecting the origins and destinations of their the basis for the development of many scientific
spatial movements while the aggregate models analyze disciplines and technology leading to the prosperity of
total number of flows between analysis zones. Since man's society, we must note that it cannot deal effectively
disaggregate models work at individual level, proponents with passengers', dispatchers' or drivers' feelings of
of such models claim that the data requirement for uncertainty, vagueness and ambiguity. Since the fuzzy
calibration of these types of models may be significantly set theory recognizes the vague boundary that exists in
lower (Ruiter and Ben-Akiva, 1978). Even though there some sets, different fuzzy set theory techniques need to
have been many alternative formulations for the be used in order to properly model traffic and
aggregate trip distribution models (namely, growth factor, transportation problems characterized by ambiguity,
fratar, intervening opportunities, gravity or regression subjectivity and uncertainty (Teodorovic, 1999).
models), the gravity model is the most preferred one over A fuzzy logic system is a nonlinear system that maps a
the years despite all of its drawbacks (Murat and Celik, crisp input vector into a crisp scalar output. When solving
2010). a large number of different traffic and transportation
An analysis of the factors influencing traveler behavior problems, this is what we actually do: map a crisp input
and destination selection shows that the attractiveness of vector into a crisp scalar output. Fuzzy logic could be
the traffic zone strongly affects the trip generation volume used successfully to model situations in which people
(Yao Liya et al, 2008). Different studies have been make decisions in complex environments that are very
considered the impact of social variables and private hard to develop a mathematical model. Such situations
information on trip generation (Smiller and Hoel, 2006). for example often occur in the field of traffic and
But none of the researches mapped generation variables transportation when studying the work of dispatchers or
of the origin and attraction variables of destination to final modeling choice problems. Present experience shows
trip flows between origin-destination. Due to the that there is room for the development of different
stochastic nature of traffic flow and the strongly non- approximate reasoning algorithms when solving complex
linear characteristics of traffic dynamics, methods of soft problems of this type (Teodorovic, 1999). Fuzzy set
computing have received much attention since early 90s theory has a successful background in solving trans-
and considered as alternatives for the traditional portation problems that is presented in the works of Chen
statistical models (Celikoglu and Cigizoglu, 2007). Among et al. (1990), Lotan and Koutsopoulos (1993a,b), Xu and
these methods the artificial neural network (ANN) have Chan, 1993), Teodorovic and Babic (1993), Chang and
been commonly applied in a number of areas of transport Shyu (1993), Chanas et al. (1993), Deb (1993), Nanda
(Dougherty, 1995), including the studies of traffic volume and Kikuchi (1993), Vukadinovic and Teodorovic (1994),
forecasting (Yun et. al. 1998), short-term traffic flow Teodorovic et al. (1994), Teodorovic and Kalic (1995),
prediction (Chen and Muller, 2001, Messai et. al., 2002), Milosavljevic et al. (1996).
macroscopic modeling of freeway traffic (Zhang et al., In this paper we have tried to propose a framework to
Jassbi et al. 507

Figure 1. Conceptual model of trip forecasting input-output space mapping.

develop transportation experts’ subjective pattern based PROBLEM DEFINITION
on generation and attraction variables of a given origin-
destination pair. Meanwhile, fuzzy inference system The final goal of any transportation planning method is to
approach is used to map these variables to total number forecast the number of trips between two given region of
of trips between origin-destination pair. under studied zone. Based on the forecasted trips
between two given regions, transportation infrastructures
(such as bridges, highways, tunnels, subways, etc.) are
Novelties of the proposed approach designed to serve the demands in predefined level of
service. This problem could be interpreted as a function
The novelties of the proposed approach could be approximation problem in which we aim to map input
summarized as follows: space to output space where:
Input space is ‘Trip Generation variables and trip
1. Moving from computation efficiency to decreasing attraction variables’. These are the variables that have
forecasting error could be pronounced as a dilemma in direct impacts on the generation and attraction of trips of
decision making and forecasting problems and also could any given region. These variables have normally the
be translated as a trade off problem. The complex prob- social and economical nature and are common variables
lems challenge decision makers to define the transfer that are used in transportation planning studies. Table 1
function more precisely with more variables. Although this shows the variables that are used in Tehran transporta-
could help to define the system but, calculations of tion planning program. Output space is ‘the number of
dynamic systems in simulation process would increase trips between two given regions’. The number of trips
exponentially. Modeling traffic flows and trip forecasting is would be approximated by a transfer function that is
a complex problem that applying fuzzy concept in this formed based on the relations of trip generation
field could be helpful because of its capabilities in capabilities of origin and trip attraction capacities of
modeling complex system in none computationally destination.
expensive manner.
2. Although there are different forecasting models exis-
ting in modeling traffic flows and trip forecasting between Conceptual model
metropolitan regions, but most of them are hard models.
This fact most be noted that traveling between two points Figure 1 illustrates the conceptual framework of the trip
in a city (that could be generalized to two regions in a forecasting problem. The main concern of the problem is
metropolis) is a human behavior that could be adopted to approximate the transfer function that is capable of
from human perception to overcome his/her needs. Thus, mapping trip attraction and generation variables to
applying a fuzzy concept that maps human subjective number of trips between two given regions. As it
patterns to mathematical models could be resulted in addressed in introduction section, Different kinds of
more meaningful results. Due to our best knowledge the transfer functions had been used in this problem. Here, a
combination of the fuzzy proposed model has never been hybrid fuzzy inference system is used to map input space
reported in literature before. to output space.
508 Afr. J. Bus. Manage.

Table 1. Variables with impacts on trip generation and attraction based on the goal of the trip.

Trip goals Generation Attraction
Occupation in the residence area Occupation in the working area
Population Business/administrative/agricultural/industrial
Residential space Land space
Number of residential buildings
Work trips Business/administrative/agricultural/industrial
Population density
building space
Household number
Average of automobile per household Number of Business/administrative/
Average price of the one square meter of the land agricultural/industrial units

Educational Trips Population
Schools’ spaces
Population of the students
Car Ownership Number of schools/ number of students/
Residential space number of schools classes.
Traffic zone space Number of universities/ number of universities
Average of automobile per household students

Business land space
Occupation in the residence area
Household numbers
Shopping trips Occupation in the working area
Residential space
Car Ownership Business land use space
Residential Space Land/building price Number of business units

Population Number of retail stores
Number of residential buildings Number and capacity of cinemas/ mosques/
Social trips Car Ownership exhibition centers/ parks/ hospitals
Social land use space
Distance to entertainment complexes

Trips that are not started
All generation variables All attraction variables
from or ended to home

Fuzzy logic and fuzzy inference systems
There are two types of fuzzy inference systems that
Fuzzy logic is a form of multi-valued logic derived from usually can be implemented: Mamdani-type and Sugeno-
fuzzy set theory to deal with reasoning that is type. These two types of inference systems vary
approximate rather than precise. In contrast with "crisp somewhat in the way outputs are determined. Fuzzy
logic", where binary sets have binary logic, fuzzy logic inference systems have been successfully applied in
variables may have a truth value that ranges between 0 fields such as automatic control, data classification,
and 1 and is not constrained to the two truth values of decision analysis, expert systems, and computer vision.
classic propositional logic (Novák et al., 1999). Fuzzy Because of its multidisciplinary nature, fuzzy inference
inference is the process of formulating the mapping from systems are associated with a number of names, such as
a given input to an output using fuzzy logic. The mapping fuzzy-rule-based systems, fuzzy expert systems, fuzzy
then provides a basis from which decisions can be made, modeling, fuzzy associative memory, fuzzy logic
or patterns discerned. The process of fuzzy inference controllers, and simply (and ambiguously) fuzzy systems.
involves: The input space variables could be summarized as
Tables 2 and 3.
i. Defining If-Then rules Mamdani's fuzzy inference method is the most
ii. Defining membership functions commonly seen fuzzy methodology. Mamdani's method
iii. Appling logical operations was among the first control systems built using fuzzy set
Jassbi et al. 509

Table 2. Generation variables (input space variables).

Input space variables Generation variables
Occupation in the residence area
Residential space
Population density
Number of households
Car ownership rate
Average price of one square meter of land
Students population
Traffic zone space
Number of residential buildings
Distance to entertainment complexes
Adapted from Tehran comprehensive transportation and traffic studies Co. (TCTTS).

Table 3. Attraction variables (input space variables).

Input space variables Attraction variables
Occupation in the working area
Business/ Administrative/Agricultural/Industrial Land Space
Administrative building space
Number of Administrative/Business/Industrial Buildings
Schools’ space
Number of Students/Schools/Classes
Number of Universities/Students
Number of Retailers
Number and Capacity of Cinemas/Mosques/ Exhibitions/
Sport and Social Centers Spaces
Adapted from Tehran comprehensive transportation and traffic studies Co. (TCTTS). (

theory. It was proposed in 1975 by Ibrahim Mamdani that revealed the impacts of the generation and attraction variables
(Mamdani 1975) as an attempt to control a steam engine on trip distributions.
iii. Transportation planning procedures in metropolises (Tehran
and boiler combination by synthesizing a set of linguistic Comprehensive Traffic Study Company).
control rules obtained from experienced human
operators. As the most part of fuzzy logic are common To map input space to output space, different fuzzy scenarios could
knowledge now, readers are referenced to bibliography be considered. Figure 2 illustrates the framework of conceptual
(Jang 1997, Mamdani 1975, Sugeno 1985). fuzzy model that is used to forecast the number of trips in this work.
As it illustrated in Figure 2, three independent fuzzy inference sys-
tems are used. The first FIS is aimed to map generation variables to
PROBLEM FORMULATION number of generated trips . The second FIS is designed to map
attraction variables to number of attracted trips . At the end, the
In order to forecast the trips based on fuzzy inference approach, a third FIS is designed to combine the output of the first and second
fuzzy rule base must be developed. The relations between varia- FIS.
bles would form the fuzzy rule base. These relations have been
adapted from:
i. Transportation experts’ opinions about the impacts of the input
space variables on the number of trips As it is shown in Figure 2, this scenario consists of three
ii. Literature review: Past studies including soft and hard approaches phases in which each phase has its own FIS.
510 Afr. J. Bus. Manage.

Figure 2. Conceptual fuzzy model: Mapping trip generation and attraction variables to total number of trips between
two given regions.

Phase I “If is low and then is medium”

Mapping trip generation variables as input space to total In fact, the whole fuzzy system is an exploration of
number of generated trips by the given region ( ), as expert’s subjective pattern that maps input space to
the output space using a proper FIS as the function output space. Input space consists of 11 demographic
approximator. variables ( ) with direct impacts on trip generation that
are presented in Table 2. The designed FIS for this
phase is a Mamdani type fuzzy inference system as
Phase II illustrated in Figure 3.

Mapping trip attraction variables as input space to total
number of attracted trips by the given region ( ), as the Phase II
output space using a proper FIS as the function
approximator. The structure of modeling in this phase is similar to
Phase I. Input space consists of 10 demographic
variables ( ) with direct impacts on trip attraction of the
Phase III exact region that are presented in Table 3.
The designed FIS for this phase is a Mamdani type
Aggregation of and for two given regions and fuzzy inference system as shown in Figure 4. Similar to
calculating the final total number of trips between two phase I rule generation procedure, fuzzy rule base in this
regions ( ). Every phase has its own FIS that maps phase is shaped based on transportation expert’s
opinions. After aggregation of expert’s opinions and
input space to output space. As it shown in the problem
eliminating rules overlaps, 39 rules are selected. Now, we
definition section, the final solution of the problem would
pass through three phases. The modeling procedure is: have two different FIS that map demographic variables to
trip attraction and generation conditions of two given
regions, respectively. In order to forecast the trip flows
Phase I
between an OD pair, the output of the previous phases
have to be combined in a new FIS.
The initial fuzzy rules and membership functions types
and shapes are adjusted based on experts’ opinions. In
order to catch experts’ subjective pattern about the Phase III
impacts of generation variables on number of generated
trips, a proper fuzzy questionnaire is designed to ask Based on Phases I and II, the trip generation condition of
about the relations. The results are aggregated and 70 the origin and trip attraction condition of the destination
initial rules are generated. Enjoying rules reduction are known. Thus, we need to design another FIS to map
techniques, number of rules has been reduced to 48. The trip attraction condition, as input space to trip generation
structure of rules is for example as follows: condition, as output space. Figure 5 shows this procedure.
Jassbi et al. 511

Figure 3. Phase I fuzzy inference system.

Figure 4. Phase II Fuzzy Inference System.

this procedure. This fuzzy inference system accepts the output of Phase II fuzzy inference system. Rule base
outputs of Phases I and II as its inputs. Here, Tproduct is of this fuzzy inference system consists of 11 rules that
the output of Phase I fuzzy inference system and Tattract is are adapted from transportation experts. The output of
512 Afr. J. Bus. Manage.

Figure 5. Phase III fuzzy inference system.

Figure 6. Surface of phase III FIS rules.

the model is the number of trips between two given Conclusion
regions that is represented by Tnum. Figure 6 visualized
the surface of number of trips, as the output versus trip Trip, as defined, “to move between origin to destination
attraction condition and trip generation condition, as the by mode on route in an exact period of time”, is
input. completely a human choice that is made based on
The three phases model, maps demographical conditions. Here we defined the condition as the
variables to number of trips. In order to forecast the combination of variables with impacts on reasons of the
number of trips between two given regions, demographic trips. The reasons are social variables that made
variables must be fed to the model and result would be commuters to move from exact location to another one in
the number of trips. Figure 7 shows the block diagram of a metropolitan area. These variables are, operative in trip
continuous simulated framework of the model. generation and trip attraction of regions. As it mentioned
Jassbi et al. 513

Figure 7. Proposed fuzzy Simulink™ model.

in introduction section of the paper, the key to successful Golob FT (2000). A simultaneous model of household activity
forecasting of the future is to recognize the existing participation and trip chain generation. In: Transportation Research
Record, J. Transport. Res. Board., No. 34, TRB. Washington, USA.
patterns correctly. In fact we need functions to help us 355-376.
discover existing patterns of trips between regions. Jang JSR (1997). Sun, Neuro-Fuzzy and Soft Computing: A
Because of fuzzy nature of human decision making Computational Approach to Learning and Machine Intelligence,
Prentice Hall.
process, it seems that phenomena that are related to
Lin CT, Chiu H, Chu PY (2006). Agility index in supply chain. Int. J.
human choice could be explained using fuzzy logic more Prod. Econ. 100: 285-299.
accurately. In this paper a logical framework is proposed Lotan T, Koutsopoulos H, (1993). Route choice in the presence of
based on fuzzy inference systems to approximate trip information using concepts from fuzzy control and approximate
reasoning. Transp. Plan. Technol., 17: 113-126.
distribution function.
Lotan T, Koutsopoulos H (1993). Models for route choice behaviour in
the presence of information using concepts from fuzzy set theory and
approximate reasoning. Transport., 20: 129-155.
Makvandi P, Alavi, SH, Hajiha A (2006). An Exploration of Experts'
Subjective Patterns in Behavioral Based Job Qualification Using
Choquet Integral, Proceedings of the 6th WSEAS Int. Conf. on
Celikoglu B, Hilmi Cigizoglu K, Hekmet (2007). Public transportation trip Systems Theory, Scientific Computation, Elounda, Greece, August
flow modeling with generalized regression neural networks. Adv. Eng. 21-23. 14-18.
Softw., 38: 71–79. Mamdani EH, Assilian S (1975). An experiment in linguistic synthesis
Chanas S, Delgado M, Verdegay JL, Vila MA (1993). Interval and fuzzy with a fuzzy logic controller." Int. J. Man-Machine Stud., 7(1): 1-13.
extensions of classical transportation problems. Transp. Plann. Messai N, Thomas P, Lefebvre D, El Moudni A (2002). Optimal neural
Technol. 17: 203-218. networks architectures for the flow-density relationships of traffic
Chang YH, Shyu TH (1993). Traffic signal installation by the expert models. Math. Comput Simul., 60(3–5) 401–409.
system using fuzzy set theory for inexact reasoning. Transp. Plan. Milosavljevic N, Teodorovic D, Papic V, Pavkovic G (1996). A fuzzy
Technol., 17: 191-202. approach to the vehicle assignment problem. Transp. Plan. Technol.,
Chen H, Muller SG (2001). Use of sequential learning for short-term 20:33-47.
traffic flow forecasting. Transp. Res. Part C Emerg. Technol., 9(5): Murat H, Celik S (2010). Sample size needed for calibrating trip
319–336. distribution and behavior of the gravity model. J. Transp. Geogr., 18:
Chen L, May A, Auslander D (1990). Freeway ramp control using fuzzy 83–190.
set theory for inexact reasoning. Transp. Res., 24A: 15-25. Nanda R, Kikuchi S (1993). Estimation of trip O-D matrix when input
Deb SK (1993). Fuzzy set approach in mass transit mode choice. In: and output are fuzzy. In: Ayyub, B.M. (Ed.), Proceedings of ISUMA
Ayyub, B.M. (Ed.), Proceedings of ISUMA '93, Second International '93, Second International Symposium on Uncertainty Modeling and
Symposium on Uncertainty Modeling and Analysis. IEEE Computer Analysis. IEEE Computer Press, College Park, Maryland. pp.104-
Press, College Park, Maryland, pp.262-268. 111.
Dougherty MS (1995). A review of neural networks applied to transport. Novák V, Perfilieva I, Močkoř J (1999) Mathematical principles of fuzzy
Transp Res Part C Emerg Technol., 3(4): 247–260. logic Dodrecht: Kluwer Academic. ISBN 0-7923-8595-0.
514 Afr. J. Bus. Manage.

Ortuzar JD, Willumsen LG (2001). The Traffic Assignment Problem: Teodorovic D, Kalic, M (1995). A fuzzy route choice model for air
Models and Methods. VSP, Wiley, New York. Utrecht, The transportation networks. Transp. Plan. Technol., 19: 109-119.
Netherlands. Vukadinovic K, Teodorovic D (1994). A fuzzy approach to the vessel
Ross TJ (2004). Fuzzy Logic with Engineering Applications. Second dispatching problem. Eur. J. Oper. Res. 76: 155-164.
Edition. John Wiley and Sons Ltd. The Atrium, Southern Gate, Xu W, Chan Y (1993). Estimating an origin-destination matrix with fuzzy
Chichester, West Sussex PO19 8SQ, England. weights. Part 1: Methodology. Transport. Plan. Technol., 17: 127-
Ruiter ER, Ben-Akiva ME (1978). Disaggregate travel demand models 144.
for the San Francisco area: system structure, component models and Yao L, Guan H, Yan H (2008). Trip Generation Model Based on
application procedures. Transp. Res. Record, 673: 121–128. Destination Attractiveness, Tsinghua Sci. Technol., 13(5): 632-635.
Smiller J, Hoel LA (2006). Assessing the utility of private information in Yun SY, Namkoong S, Rho JH, Shin SW, Choi JU (1998). A
transportation planning studies: A case study of trip generation performance evaluation of neural network models in traffic volume
analysis. J. Socio-Econ. Plan. Sci., 40(3): 94-118. forecasting. Math. Comput Model. 27(9–11): 293–310.
Sugeno M (1985). Industrial applications of fuzzy control. Elsevier Zhang H, Ritchie SG, Lo ZP (1997). Macroscopic modeling of freeway
Science Pub. Co. traffic using an artificial neural network. Transp. Res. Rec.,
Susilo YO, Kitamura R (2007). Structural changes in commuters’ daily 1588:110–119.
travel: the case of auto and transit commuters in the Osaka
metropolitan area of Japan, 1980–2000. Transp. Res. Part A: Policy
Pract. 42: 95–115.
Teodorovic D (1999). Fuzzy logic systems for transportation
engineering: the state of the art, Transport. Res. Part A 33: 337-364
Teodorovic D, Babic O (1993). Fuzzy inference approach to the flow
management problem in air traffic control. Transp. Plan. Technol., 17:
Reproduced with permission of the copyright owner. Further reproduction prohibited without