RISK RANKING OF ROAD SECTIONS ON HIGHWAYS USING ORDERED WEIGHT AVERAGING (OWA) DECISION RULE Yaowaret Jantakat 1 , Sunya Sarapirome 2 , Suwit Ongsomwang 3 , Patiwat Littidej 4 School of Remote Sensing, Institute of Science, Suranaree University of Technology, Thailand Tel. (66) 044-224652 and Fax. (66) 044-224316 e-mail 1 : st102185@yahoo.com e-mail 2 : sanyas@g.sut.ac.th KEY WORDS: road risk ranking, multi-criteria decision analysis, ordered weight averaging, GIS ABSTRACT: Ranked road sections in terms of risk together with ranked weights of factors considered to cause accident for each section are highly effectual information for road safety implementing planning. To achieve this goal, 36 road sections from 5 highways in Nakhon Ratchasima, Thailand with varying slopes, surface widths, and a number of connection routes, initially selected from 166 sections by using 3-year accident data, are ranked into order from the highest risk to the lowest risk using ordered weight averaging (OWA) decision rule. OWA is a multi-criteria evaluation procedure using combination operators. Apart from risk ranking of road sections, the result shows that slope is considered to be the highest rank among risk factors for 16 sections, while 15 and 5 sections are for the number of connection routes and surface width respectively. In addition, a number of ranking sections within each certain highway can be obtained i.e. 17, 12, 3, 2, and 2 sections for highway no. 205, 207, 208, 2150, and 2160, respectively. Then, the priority of road safety programs to reduce accident rate can be set effectively. 1. INTRODUCTION 1.1 Backgrounds As a consequence of road accident, the problem of fatalities and injuries is currently acknowledged to be an uncommonly global phenomenon with authorities virtually in all countries of the world that continue aggravating in the number of deaths and serious injuries on their roads. Furthermore, WHO (2004) stated that nearly 1.2 million fatalities and over 20 million injured in the accidents of road each year while every day thousands of people are killed and injured on world’s roads. Then, determining the relevant risk factors in a given situation is one of the prime tasks for transport designers, regulators and policy makers are to minimize the total risk in such situation (Turner et al., 2006). In addition, a technique of multi- criteria analysis is one important approach for development of road safety strategies (AUSTROAD, 2000). Consequently, OWA is selected for ranking risk factors of road section due to the strong benefits of the OWA method for interactive features; dynamic map updates and linked displays by allowing users to change the parameters, re-evaluate criteria and visualize new rankings. OWA is a multi-criteria evaluation procedure using combination operators; is weighted sum with ordered evaluation criteria. The task of OWA, moreover, importantly involves two sets of weights: criterion importance weights and order weights (Yager, 1988). 1.2 Objectives In this study, there are two main objectives: 1) road section ranking and 2) ranking of criteria in each road section.