Return to workshop home page

On the Mathematical DEA-MCDM Relationship

Per J. Agrell

This presentation solidifies the bounds between the non-parametric data envelopment analysis (DEA) models for efficiency analysis and multi-criteria decision making models for the linear and non-linear case. By drawing on the properties of a partial Lagrangean relaxation, a correspondence is shown between the CCR, BCC and free disposable hull (FDH) models in DEA and the MCDM model. One of the implications is a characterization that verifies the sufficiency of the weighted scalarizing function, even for the non-convex case FDH. Thus, an input/output-oriented model is shown to be equivalent to a maximization of the weighted input/output, subject to production space feasibility. The discussion extends to the recent developments: the free replicability hull (FRH), an intermediate form ERH and the non-convex models by Petersen (1990). The FRH model is shown to be a true MIP, whereas the latter can be characterized as the CCR and BCC models.

Return to presentations       Return to workshop program

Multicriterial Ranking Methods for Evaluating Bank Branch Performance

Fuad Aleskerov
Hasan Ersel
Reha Yolalan

For evaluating the performance of bank branches, 15 multicriterial ranking methods are suggested. The methods are classified into 6 groups: methods using upper and lower contour sets, methods using tournement matrices, methods using value function, q-paretian procedures, procedures using approximation of pareto relation, and methods using data envelopment analysis. For the first five groups of methods, to make the results more robust with respect to deviations in the data the insensitivity threshold concept is introduced. It allows to compare not point-wise evaluations of alternatives with respect to multiple criteria, but compare intervals in which unknown evaluations are set up. The methods are explained via an illustrative example, and are applied to real life data for retail bank branches of a large commercial bank in Turkey. The obtained rankings are compared using Hemming’s distance and pseudo-metric. A software has been developed to implement these methods. Two versions of software are in use: with 500 alternatives and 30 criteria, and with 800 alternatives and 10 criteria.

Keywords: Multicriterial Ranking Methods; Insensitivity Threshold; Bank Branches; Bank Performance

Return to presentations       Return to workshop program

Group Decisions in the Computer Supported Collaborative Design (CSCD) Using AHP

Nigan Bayazit
Pinar Yalçin
Orhun Berkün

The objective of this paper is to shed some light to the problems and benefits of Computer Supported Cooperative Design Decisions. Different role relationships in the design team create a different kind of correspondence between the members of the design team. The role of group in decision making is the main subject of this paper. AHP can be used as a decision support program for design stages during the design process.

Analytic hierarchy process is developed by Thomas L. Saaty who introduced a method of assigning weights to categories based on paired comparisons. In this process the choices depend on the individuals' expertise and the group relations. Therefore, the logic of choice does not always depend on the descriptive qualities of the computer program. Passing messages and information while working in a group in front of a computer is different from the working around a table in a face to face decision making group.

Computer Supported Cooperative Design issues is important in educational and professional context. Organizational parameters of the computer network also influence the types of correspondence on design problems. These parameters are the number of the participants on a design problem, the hierarchical system of the organization, the job structure and stages of the design problem.

In front of a computer, communication channels influence the learning process for socialization and learning processes of cognitive understanding. These differences in the learning process need further investigations.

Return to presentations       Return to workshop program

Measurement and Evaluation of Store Performance Based on Multiple Criteria

Altan Cöner

This study is based on a field research conducted at 20 large size retail stores and 5 department stores in Istanbul. The field research and the empirical study conducted covered all facets of store operations such as location, store windows, size, effectiveness of sales force, merchandizing, in-store promotions, store interior design, inventory availability and selection, pricing policies, customer service and after sales service. The performance of the stores based upon such multiple criteria was evaluated and points were assigned for each criterion. Finally the overall performance scores were compared to the firm's financial performance. It is observed that there is a high correlation between performance scores and financial performance/profitability.

Return to presentations       Return to workshop program

Measuring Relative Financial Efficiency of the Organizations

Ahmet Burak Emel
Muhittin Oral
Reha Yolalan

This study proposes a method for financial performance analysis. Traditional financial ratio analysis examines one ratio at a time. In this study, to enrich the unidimensional nature of ratio analysis, Data Envelopment Analysis (DEA) is used to measure relative financial efficiencies of organizations via ratios computed from their financial statements. DEA allows decision-makers to make a multiple-criterion financial analysis as well as to integrate overall performance into a single financial efficiency score. An empirical study was conducted to analyse the performance of a group of companies in a selected sector. To identify the financial ratios to be used in DEA, first a large set of financial ratios are scanned and grouped using factor analysis. Variables, selected from different principal factors, are used in the scoring function in order to consider the multidimensional structure of the financial performance. According to their efficiency scores, organizations are classified as "good" or "bad". It is observed that for "good" firms; liquidity, capital adequacy and profitability is higher.

Keywords: Performance Evaluation; Data Envelopment Analysis; Financial Ratio Analysis

Return to presentations       Return to workshop program

A Multiple Criteria Approach for the Evaluation of the Rail Transit Networks in Istanbul

Haluk Gerçek
Birsen Karpak
Tülay Kilinçaslan

In this paper, rail transit projects are evaluated based on financial, economic, system planning, policy, and environmental impacts criteria for the improvement of the public transportation system in the European side of Istanbul. Analytic Hierarchy Process (AHP), a multiple criteria decision support system, is used to assist The Greater Istanbul Municipality (GIM) which aims to extend the railway systems in order to improve public transportation of the city. The municipality has already built and put into operation a fully grade separated 18 km-long light rail transit (LRT) system in the European side of the city. A subway line is under construction; but, under the pressure of increasing travel demand in the area, new extension lines have become necessary. In 1995, the GIM commissioned a consortium, Tekfen (Turkey) and Delcan (Canada), to carry out a strategic network planning study to identify the best rail transit projects to improve the public transportation system in the European side of the city. This paper presents the evaluation process of the rail network alternatives developed by Tekfen and Delcan to provide a solid base for the final decision made by the GIM. Alternative rail networks are evaluated according to financial, economic, systems planning, policy, and environmental impact criteria by using Analytic Hierarchy Process (AHP). Two land use scenarios have been developed in order to predict the future travel demand in the horizon year of 2010. A conventional transportation model was used to predict the travel demand that will likely occur on the transportation system as a result of the future land use and population changes.

Return to presentations       Return to workshop program

Decision Making Tariff Based on Marginal Cost for Turkish Electrical System Related With Load Research and Demand Management

H. Gönül Günvaran

Modern societies have become increasingly dependent on various types of energy sources, among which electric power has occupied a dominant position. Besides, electricity is the most important infrastructure sector in every country and its investment necessitates large amount of capital. Because of the above reasons, it is. There are two tariff calculating methods which are based on embedded cost and marginal cost. Tariff based on embedded cost is related to the past expenditures while marginal cost tariff is based on the future expenditures. The tariff based oh marginal cost has the following properties which are so important for the electrical system. Obvious the importance of decision the correct power pricing policy in this sector. Tariff should be maximize the net economic benefit of electricity consumption to society. It should also be fair while allocating costs among consumer groups and power prices should raise sufficient revenues to meet the financial requirements of the sector. Because of the reasons mentioned above, tariff based on marginal costs has been taken into account and marginal cost components have been calculated for Turkish Electrical system. While marginal capacity cost of generation is calculated analytically, for the marginal energy cost calculation, by using Baleriaux-Booth Technique "Equivalent Load Curve Probabilistic Simulation Model" has been developed. This is short-run planning model. For the marginal cost capacity costs of transmission and distribution network, the average incremental cost technique is used. The above techniques are implemented to Turkish Electrical System. It had been tried to be proved that Turkey should pass tariff structure based on marginal cost.

Keywords: Electrical Tariff; Generation; Transmission; Distribution; LRMC; SRMC

Return to presentations       Return to workshop program

Resource Allocation for Urban Quality of Life in Istanbul: Inhabitants' Perspective

Ümit Güvenç
Burç Ülengin
Füsun Ülengin

The purpose of this paper is to model the desires, expectations and priorities of the inhabitants of Istanbul from a multidimensional perspective. In this way, effective allocation of the city’s resources can be achieved to improve the quality of life for such a large number of people, which is the primary concern of the local authorities as well as the urban planners. A survey is conducted in Istanbul so that the priorities of the inhabitants are revealed and the city where they would like to live is portrayed. The data obtained are used as input for hierarchical conjoint analysis. The survey is primarily based on the evaluation of hypothetical, orthogonally-designed city profiles for four different constructs and a bridging construct on a 0-10 rating scale. The relative importances of the constucts and their attributes are estimated at both the individual and the aggregate level. A segmentation is made based on the demographic and social characteristics of the respondents to reflect different classes. The research is an interdisciplinary group work acting as a bridge between urban planning and multiattribute decision making, thus judgments of experts from different disciplines are used in every stage of the study.

Keywords: Hierarchical Conjoint Analysis, Urban Quality of Life, Resource Allocation, Segmentation

Return to presentations       Return to workshop program

Global Decision Support by Web-HIPRE - A Java-applet for Value Tree and AHP Analysis

Raimo P. Hämäläinen
Jyri Mustajoki

Web-HIPRE is a Java-applet for multiple criteria decision making based on the well-known decision support software HIPRE. Being located in the Internet, Web-HIPRE can be accessed from everywhere in the world. This opens up a completely new era and dimension in decision support. Web-HIPRE provides a common platform for individual and group decision support. The individual models can be processed at the same or at different times and the results can be easily shared. An essential web feature is the possibility to define links to other Internet-addresses. These links can refer to graphical or any other kind of information such as sound or video describing the criteria or alternatives. This can improve the quality of decision support dramatically. Web-HIPRE supports several weighting methods including AHP, SMART, SWING, SMARTER and value functions. The results are shown by bar graphs and the sensitivity analysis. Web-HIPRE supports also the use of regular HIPRE models. The on-line use of Web-HIPRE will be demostrated by illustrative examples. Web-HIPRE is located in http://www.hipre.hut.fi/.

Keywords: World Wide Web; Decision Analysis; Value Tree; AHP; Multimedia

Return to presentations       Return to workshop program

The Impact of Political Risk on the Rankings of Emerging Markets: An Analytical Assessment

Birsen Karpak
Erdogan Kumcu

A number of worldwide trends are making a group of economies prime targets for the US companies-- the so-called emerging countries. These markets have become attractive and promising as destinations for technology and goods as well as a pool for offshore sourcing. Emerging Markets are high-growth developing countries that represent attractive business opportunities for Western firms. The US Department of Commerce lists 10 "Big Emerging Markets": the Chinese Economic Area (China, Hong Kong, Taiwan), ASEAN (Indonesia, Malaysia, The Philippines, Singapore, Thailand, Brunei), India, South Korea, Argentina, Mexico, Brazil, Poland, Turkey, and South Africa. The Economist adds Chile, Venezuela, Greece, Israel, Portugal, the Czech Republic, Hungary, and Russia.

The emerging markets (EMs) share remarkable features in terms of economic potential. Some are large geographically. Demographically they have a young labor force. They are attractive for both selling and sourcing, enjoy outstanding growth rates, and are prospects for market expansion.

We have ranked 23 emerging markets using the Analytic Hierarchy Process (AHP) as the decision analysis methodology and ECPro as the software. We used 13 economic, political, and social variables to characterize a market's attractiveness from the viewpoint of US management. They are market size, market growth rate, market intensity which consists of purchasing parity of GNP per capita and private consumption expenditure per capita, market consumption capacity, commercial infrastructure which consists of telephone mainlines per capita, paved road density, commercial vehicles per capita, population per retail outlet, and percentage of homes with TV, economic freedom index and market receptivity which consists of average annual growth rate of imports from the US over the past five years and per capita imports from the US.

The assumptions associated with the behavior of emerging markets are that conditions are highly volatile, and that social and political conditions greatly affect the economic policies under which multinational corporations will be operating. Collectively these assumptions illustrate intangible aspects of the international market place. Any model addressing EMs should be able to handle qualitative criteria related to such issues. Sensitivity analysis of the model showed that the overall ranking was not very sensitive to political risk for some countries whereas in practice sometimes this criterion is used as a sole elimination criterion. The impact of changing the importance of different criteria, adding qualitative criteria to the analysis as well as the stability of on prioritizing the EMs will be discussed.

Return to presentations       Return to workshop program

Multicriteria Approach to Total Quality Control

Çetin Kaya

This presentation will be dealing with the multicriteria measurements in the control of the quality of goods produced and the services rendered. In order to gain a competitive advantage, improving the current level of quality is now emerging as an unavoidable success factor for the enlightened organizations. Needless to say that lower costs and higher quality are the two main paths for the greater customer satisfaction. Within this context applying the multicriteria approach to the quality problems deserves more attention than ever before. That is why we will stop at every station of the quality labyrinths throughout the presentation.

Return to presentations       Return to workshop program

Analyzing and Improving Efficiency in Organizations

Pekka Korhonen

Data Envelopment Analysis (DEA), originally proposed by Charnes, Cooper and Rhodes [1978], has become a widely used method for evaluating the performance of decision making units (DMUs). DEA measures the relative efficiency of comparable DMUs performing the same task and using similar multiple inputs to produce similar multiple outputs. The purpose of DEA is to empirically estimate the efficient frontier based on the set of DMUs available for the analysis. A DMU is efficient if there is no other DMU (or a positive linear/convex combination of them) that can either produce more outputs by consuming the same amount or less inputs or produce the same amount or more outputs by consuming less or the same amount of inputs as the DMU under consideration. DEA provides the user with information about the efficient and the inefficient units, as well as about the efficiency scores and reference sets for inefficient units. The results of the DEA analysis, especially the efficiency scores, are used in practical applications as performance indicators of the DMU in question.

The DEA-approach is a nice method in identifying inefficiency, but in the analysis the decision maker’s preferences are ignored, and the DEA-approach does not help the DM to see how to improve performance. As a solution to this problem, we proposed the use of so-called Value Efficiency Analysis (Halme, Joro, Korhonen, Salo, and Wallenius, [1998]). Other ways to develop the DEA is to insert into the analysis the features which help the DM to analyze efficient resource allocation and to design new (efficient) units. In this presentation, we will give an overview on the main features of the Data Envelopment Analysis and discuss those new additional features.

Keywords: Efficiency Analysis; Data Envelopment Analysis; Value Function; Practical Application

Return to presentations       Return to tutorial program

The Additive and the Multiplicative AHP

Freerk A. Lootsma

The Analytic Hierarchy Process (AHP) of Saaty (1980) is a widely used method for Multi-Criteria Decision Analysis, presumably because the decision makers can easily understand the elicitation of preference information. The original AHP has been criticized, however, because the algorithmic steps do not properly take into account that the method is based upon ratio information. The shortcomings (rank reversal, results dependent on the order of the computations) can easily be avoided in the Additive and the Multiplicative AHP to be discussed in the presentation. The Additive AHP is the SMART procedure, with pairwise comparisons on the basis of difference information. The Multiplicative AHP, with pairwise comparisons on the basis of ratio information, is a variant of the original AHP. There is a logarithmic relationship between the two methods (logarithmic coding in psycho-physics, see the decibel scale in acoustics) that we extensively employ in the elicitation of preference information.

Return to presentations       Return to workshop program

Optimization System WWW-NIMBUS

Kaisa Miettinen
Marko Mäkelä

Finding optimal solutions to problems with several conflicting criteria often necessitates interaction with the decision maker and the solution method. Internet provides a versatile tool in realizing such interaction. Internet is easily available and sets minimal requirements to the computer facilities of the user.

We describe WWW-NIMBUS -- a pioneering interactive multiobjective optimization system in the Internet. The main principles of the implementation have been centralized computing and distributed interface. Typically, the delivery and update of any software is problematic. Limited computer capacity may also be a problem. Via the Internet, the calculations are always performed with the latest version of the system and any client computer has the capabilities of a server computer. Further, the World-Wide Web (WWW) provides a graphical user interface. No special tools, compilers or software besides a WWW browser are needed in order to utilize the system.

Return to presentations       Return to workshop program

Efficient Solutions of Discrete Multiple Objective Optimization Problems

Tom L. Morin

We consider optimal algorithms for generating the subset of efficient solutions given n outcome vectors. This has been the topic of a great deal of research in computer science. We improve on the average case performance of previous algorithms by generating a "good" starting efficient point. If that the convex hull of Y is symmetric about y1=y2=...=yp, then we can show that such a point has the maximal intersection of Y with its domination cone, is "central," and minimizes the Euclidian distance from ideal point.

Return to presentations       Return to workshop program

Computational Intelligence and Multi-objective Programming in Financial Engineering

Hirotaka Nakayama

Techniques for learning and inference in computer intelligence are expected to be effectively applied to many problems in financial engineering such as credit evaluation, portfolio selection and so on. Among them, the author has reported the importance of additional learning and forgetting in machine learning, because the environment of decision making changes over time. It is usually time ocnsuming to relearn from the beginning. Therefore, we mean by additional learning to calculate only the increment between the current status and the new one without re-learning from the beginning for revising the machine knowledge. On the other hand, since the decision rule becomes more and more complex with only additional learning, forgetting is also necessary. Here, we mean by forgetting to remove unnecessary knowledge. At first, in this paper, several techniques for additional learning and forgetting will be discussed along with applications to some problems in financial engineering.

Furthermore, we shall discuss how multi-objective programming techniques can be effectively applied to portfolio mix problems. In many transactions, the subjective judgment of investors, traders, fund managers and so on play an important role. Multi-objective programming techniques can provide a solution by incorporating the subjective judgment of human beings.

Techniques combining non-human techniques such as computer intelligence and human techniques such as multi-objective programming seem to be effective in many practical financial problems.

Keywords: Computational Intelligence; Multi-Objective Programming; Portfolio; Pattern Classification

Return to presentations       Return to workshop program

Analysis of Consumer Characteristics Which Influence the Determinants of Buying Decision by Logistic Regression Model

Rauf Nisel
Yonca Karapazar

Satisfaction of the consumer needs is the ultimate goal of business success. Hence, an effective marketing strategy must focus on serving consumers better than the competitors. The marketing manager should be interested in revealing the complexities of an individual buyer, the dynamics of the consumer behaviour, and should also try to understand the factors that the consumers show individual differences so that he/she can segment the total market in terms of those differences. Differences in decision making processes of the consumers can ease the managers to classify individuals into meaningful homogenous subgroups in that sense. Consumer decision making should be carefully studied by marketing managers to understand how information is obtained, how beliefs are formed, and what product-choice criteria are specified by customers. Products can then be developed that will fulfill the appropriate requirements of these groups. Therefore finding out the motives that lead to differences in consumers' decision making processes is a critical factor for a company to accomplish its marketing objectives in the way to satisfy its customers. Aim of this research is to identify the significant characteristics of consumers which influence the type of the motive that enables them to buy and hence better understanding and prediction of consumer buying behavior can be achieved. The research data obtained from the interview of 360 consumers randomly selected during their shopping at food stores.

Return to presentations       Return to workshop program

Decision Making with the Analytic Network Process (ANP)

Thomas Saaty

The Analytic Hierarchy Process (AHP) is a theory for decision making involving ratio scale measurement in hierarchic structures. The AHP elicits judgments from people or uses actual ratio scale measurements either directly or by interpreting existing measurements. Judgments are made on paired comparisons of elements on a common property or criterion. For accuracy, the elements compared must be homogeneous (relatively close) and if far apart they are grouped into clusters with a pivot element from one cluster to the next. The Analytic Network Process (ANP) is the generalization of the AHP to dependence and feedback among any elements involved in a decision. Thus the AHP is a special case of the ANP. This paper gives a cursory look with an illustration of the ANP. The validity of any scientific theory lies in its use in prediction. The ANP, has been applied in numerous cases of prediction with good results validated against real life information and data.

Return to presentations       Return to workshop program

Decision Support Systems in Transportation Investment: State-of-The-Art Review

Murat M. Sati

Passenger transport is a growing problem in many developing countries. Essential problem is to decide the most adequate transportation project. It is definitely important stage in transportation planning. There are some straight studies in this subject:

Two main project evaluation approaches exist: Cost-Benefit Analysis and Multicriteria Analysis. Early transportation investment policy methodology includes the application of CBA to proceed with its evaluation. This method transforms all factors into monetary value, and uses Net Present Value Method, Benefit/Cost Ratio Method or Rate of Return Method to evaluate transportation project alternatives. In addition Social cost benefit analysis method is used very intensive by transportation planners. The purpose of this methods is to measure an investment projects contribution to the economic welfare of a community.

Transportation planning is inherently multiobjective in nature. Goals or objectives can be grouped as tangible and intangible factors such as cost, political issues and so on. The multicriteria decision-making techniques can be divided into two groups. The first is based on a ranking scheme approach and the second one on a mathematical programming approach.

The purpose of this study is to point out the spectrum of decision-making approaches to support transportation investment.

Keywords: Transportation Investment; Multicriteria Analysis; Social Cost-Benefit Analysis; Decision Support System

Return to presentations       Return to workshop program

A Framework for Measuring the Quality of Discrete Representations of Efficient sets

Serpil Sayin

One major argument voiced against Vector Optimization methods is that the efficient set, which these methods aspire to find, is too complicated to be of practical use for a Decision Maker. To respond to this valid argument, finding discrete representations of the efficient set rather than the efficient set itself has been proposed as a way of solving multiobjective mathematical programming problems. A number of methods have been proposed that are capable of finding discrete representations of the efficient set.

We propose a framework in which discrete representations can be evaluated, compared to each other, and judged satisfactory or unsatisfactory by a Decision Maker. Our framework includes three attributes of quality: coverage, uniformity and cardinality. We discuss how these attributes can be quantified and measured. Our suggestions lead to implementable methods for measuring the quality of representations in the linear case. We discuss various implementation issues.

Keywords: Multiple Objective Mathematical Programming; Vector Optimization; Efficient Set; Discrete Representation

Return to presentations       Return to workshop program

Fuzzy Economics: A Framework for Quantum Theory of Information Economics

E. Cemil Tarhan

Contemporary Economics is based on the pre-20th century concepts and values that relate to what is material in essence. Today's economic environment thrives primarily on the intellectual aspect of production. Our emphasis is less quantitative and more qualitative. Wealth is measured less in numbers and more in substance, form, character and design. We prefer better to more. Choice is no less significant than cost. "Creation" is as important as "duplication". Production involves intellectual costs as well as material costs. Intellectual capital is as potent as material resources.

This paper addresses the above mentioned contemporary issues in an information based economic environment. It defines the "information economy" and proposes a new theory based on the quantized aspects of intellectual production. It develops a philosophical base for understanding the discontinuous nature of research and development, the discrete behaviour of intellectual production, and the bounds and leaps that pervade technological advances.

This new quantum theory of information economics is then contrasted with the classical concepts and regimes. Quantum versions of the classical variables and observables are introduced. The "economic quanta", the "intellectual charge" and the "economic-potential-value (EPV)" are defined. A methodology for measuring the EPV and a taxonomy for classifying the intellectual production are introduced. New analytic tools and measurement tools are also suggested, and proper formulation is provided. This paper also discusses the Uncertainty and Exclusion Principles. These principles are then compared to fuzzy dynamics of information products.

Return to presentations       Return to workshop program

World Wide Web Technology in Support of Negotiation and Communication

Jeffrey Teich
Hannele Wallenius
Jyrki Wallenius

With the advent of the mass use of the Internet involving classic use of e-mail, file transfer and now more recent use of www applications, such as home pages, chat, Internet phone, audio and video and web broadcast systems, business transactions over the net are increasing dramatically. An important, yet relatively undeveloped component for electronic commerce is the negotiation of contracts and the resolution of disputes. We discuss Web-based technology and practice in supporting business negotiations and communication. The focus of the paper is on the present and the future promises of such technology.

Keywords: Decision Support; Negotiation Modeling; World-Wide-Web; Internet Communications; Electronic Commerce; Computer-mediated Communication; Technology Diffusion

Return to presentations       Return to workshop program

Problem Structuring and Evaluation Integration: A Case Study in Transportation Planning

Füsun Ülengin
Y. Ilker Topcu
Sule Önsel

This study focuses on the integration of problem structuring with the evaluation of alternatives. The multiple criteria analysis is generally based on the latter with a hidden assumption that the problem is well defined with explicitly known criteria and alternatives. However, especially the strategic planning problems are ill-defined and, thus, the first step of the analysis must be concerned with the structuring of the problem before any careful evaluation of the alternatives is conducted.

In this paper, the water crossing infrastructure selection problem is selected as a case study. The knowledge about the criteria to be used during the evaluations is acquired from 19 experts and using the COPE software their aggregate cognitive map is constructed. Based on the cluster, domain and hieset analysis etc. included in the COPE software, the compex map is converted into a hierarchical structure with the goal at the top and the alternatives at the bottom. Finally, at the second stage of the analysis, two different approaches are proposed for the evaluation of the alternatives based on this hierarchical structure: an expert system approach versus an AHP -based evaluation approach. The two approaches are applied to solve the water crossing traffic congestion of the Bosphorus at Istanbul and the results are evaluated and compared.

Keywords: Cognitive Mapping; AHP; Expert System; Transportation Planning

Return to presentations       Return to workshop program