Multiple criteria evaluation and optimization of transportation systems

Journal of Advanced Transportation, Winter, 2009 by Jacek Zak

This special issue (Vol. 43 No. 4) of the Journal of Advanced Transportation focused on: "MULTIPLE CRITERIA EVALUATION AND OPTIMIZATION OF TRANSPORTATION SYSTEMS" contains the second set of 5 articles accepted for publication. The first set composed also of 5 papers was published in volume 43-2. All the accepted manuscripts present different methodological tools, decision making models and computational algorithms that are applied for multiple criteria evaluation and optimization of transportation systems. They provide high quality, original contributions to both theory and practice of transportation research. They deal with analysis, design, economics, engineering, operations, planning and technology of different transportation mode. They also prove how powerful the Multiple Criteria Decision Making (Aiding)--MCDM(A) methodology is.

The pioneer research in MCDM(A) dates back to 1950-s. Hundreds of papers and conference presentations on MCDM(A) have been produced since then. Many methodological tools, decision making algorithms and computer--based systems have been designed and constructed within this field. Two major methodological streams of MCDM(A), represented by the European and American Schools have been developed. Those schools are based on Outranking Relation and Multi Attribute Utility Theory, respectively. Today MCDM(A) is a mature and still dynamically developing research field. It focuses on the design and development of advanced tools and methods that provide support to the Decision Maker in solving complex decision problems in which several, usually contradictory, points of view must be taken into account. As proven by both special issues MCDM(A) is also very popular in Transportation.

37 full--length manuscripts were submitted for consideration for the two special issues. The authors of the papers represented 21 countries from 5 continents, including 15 articles from Europe, 11 articles from Asia, 7 articles from North America, 2 articles from South America and 2 articles from Africa. All the papers went through the usual stringent refereeing process of the journal. As a guest editor I was supported by my doctoral student Mr. Szymon Fierek We both handled all the papers except one, which is the paper submitted by myself. This article was processed personally by a Co-Editor--in--Chief. At this point I would like to express my great appreciation to Mr. Szymon Fierek for his enormous efforts and undisputable contribution to the success of this special issue.

The finally accepted papers, presented in this issue (Vol.43 No.4) are characterized by the features presented below.

Vania Barcellos Gouvea Campos, Rui Antonio Rodrigues Ramos and Denise de Miranda e Silva Correia from Brasil present an article entitled: Multi-Criteria Analysis Procedure for Sustainable Mobility Evaluation in Urban Areas. The authors propose a multiple--criteria procedure to evaluate passengers' sustainable mobility in urban areas. A set of indicators defined along three dimensions of sustainability (environmental, economical, and social) is proposed to evaluate different aspects/factors that have substantial impact on mobility. The authors aggregate those indicators into an overall Sustainable Mobility Index (SMI) using weighted multi-criteria combination procedure and evaluate sustainable mobility based on SMI. They use expert opinions (from Brazil) to classify the proposed indicators/attributes/criteria, define the scope and content of the SMI as well as the weights of the suggested parameters (criteria). The proposed approach is applied and validated in the city of Belo Horizonte in Brazil, inhabited by 2.24 million people. The results of computational experiments, focused on calculating the SMI are presented.

Korean researchers: Kyoung Seok Shin, Jun Hyuk Kim and Yeo Keun Kim consider in their article a hub and spoke location problem (HSLP). In the article entitled: A Two-Leveled Multiobjective Symbiotic Evolutionary Algorithm for the Hub and Spoke Location Problem the authors develop a novel two-leveled multi-objective symbiotic evolutionary (TMSEA) algorithm to solve the HSLP under multiple scenarios. The authors consider HSLP as a set of four sub-problems, including: the hub location, spoke location, spoke allocation, and customer allocation sub-problem. In TMSEA, two main processes imitating symbiotic evolution and endosymbiotic evolution are introduced to promote the diversity and convergence of solutions. The evolutionary components suitable for each sub-problem are defined. TMSEA is tested on a variety of test-bed problems and compared with existing multi-objective evolutionary algorithms. The experimental results show that TMSEA is promising in solution convergence and diversity.

D. Agusdinata, J. van der Pas, W. Walker, and V. Marchau representing the Netherlands and USA focus their research on: Multi-Criteria Analysis for Evaluating the Impacts of Intelligent Speed Adaptation. The authors claim in their article that speed is a major cause of road accidents, resulting in high level of associated casualties and costs. Thus, they presume that in-vehicle speed limiters or Intelligent Speed Adaptation (ISA) devices are promising solutions. They investigate uncertainties associated with the implementation of ISA devices and they analyze the impact of ISA devices on drivers' behavior, speed levels and resulting accident rates. The authors develop an original methodology for evaluating multiple outcomes of ISA policies (safety, emissions, throughput, and cost) based on Multiple Criteria Analysis (MCA). The constructed policies (variants) take into account a range of future demand scenarios, functional relationships for performance criteria, and user responses to ISA systems. In the policies' construction process the authors apply exploratory modeling based on computational experiments. As an illustration of the proposed approach, different ISA policies are tested and evaluated on two groups of drivers. The results of computational experiments prove that when compliance with ISA policies is expected to be low, a policy aimed only at novice drivers outperforms other ISA policies on safety improvement.