Abstract—Evacuees’ behaviors have a significant impact on evacuation efficiency. Evacuation clearance time is one of the key indicators in the evacuation planning and management. Evacuees’ departure time choice, destination choice and route choice behavior based on real-time traffic information are three crucial factors used to estimate evacuation clearance time. In this paper, we use the orthogonal experimental design method to determine values of these three factors simultaneously, which can reduce evacuation clearance time significantly. Once we obtain these values, we can guide evacuees’ departure time choice, destination choice and route choice behavior to approximate these values and evacuation efficiency will improve greatly. The dynamic traffic evacuation process is modeled using a rule-based multi-agent microscopic traffic simulation system called TransWorld. The simulation experiment results illustrate the effectiveness and reliability of the proposed method. The proposed methodology, computational results and discussions can be used for future emergency evacuation planning and management. I. INTRODUCTION Emergency traffic evacuation is an effective strategy to mitigate damage of man-made or natural disasters. With the increasing size and frequency of disasters, studies of emergency traffic modeling and control strategies have become important research areas. In particular, the evacuee behavior related evacuation modeling has been attracting a lot of researcher’s attention, because evacuee behavior has a significant impact on emergency evacuation planning, management and control. Stern and Sinuany-Stern were believed to firstly develop a simulation model incorporating behavioral factors including the diffusion time of the evacuation instructions and *This work was supported in part by National Natural Science Foundation of China projects 70890084, 60921061, 90920305, 90924302, 60974095, 60904057, 61101220, 61004090, 61174172, 61104054; CAS projects 2F09N05, 2F09N06, 2F10E08, 2F11D03 and 2F11D01; National High Technology Research and Development Program of China grant 2011AA110502. Lv Yisheng is with the State Key Laboratory of Management and Control for Complex Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China (phone: 86-10-82613047; e-mail: yisheng.lv@ia.ac.cn). He is also with Dongguan Research Institute of CASIA, Cloud Computing Center, Chinese Academy of Sciences, Songshan Lake, Dongguan 523808, China. Zhu Fenghua, Chen Songhang, Ye Peijun are with the State Key Laboratory of Management and Control for Complex Systems, Institute of Automation, Chinese Academy of Sciences, Beijing 100190, China. They are also with Dongguan Research Institute of CASIA, Cloud Computing Center, Chinese Academy of Sciences, Songshan Lake, Dongguan 523808, China. Miao Qinghai is with the Graduate University of Chinese Academy of Sciences, Beijing, China. individual’s evacuation decision time in emergency planning [1]. In the analysis of the impact of household decisions on evacuation in Hurricane Floyd, Dow and Cutter considered the timing of departure and the role of information in the selection of specific evacuation routes [2]. Murray-Tuite developed one linear integer programming model to describe a family’s meeting location selection process and another one linear integer programming model to assign trip chains for drivers to pick up family members who may not have access to vehicles [3]. Stopher et al. and Alsnih et al. used multinomial and mixed logit models to determine when a household would evacuate due to bush fires, respectively [4]-[5]. Lazo et al. presented the stated-choice valuation method to study households’ evacuation decision [6]. Chiu et al. developed a real-time traffic management system for evacuation, in which they considered evacuee behavior responses to management strategies [7]. Further, Chiu et al. proposed a behavior-robust feedback information routing strategy to improve evacuation efficiency [8]. Li et al. simulated pedestrian evacuation using Vissim, where they considered the spatial distribution of pedestrians [9]. Hu et al. proposed a minimum-safety-distance-based evacuation car-following model based on the Gipps car-following model, in which they incorporate driver mental and behavioral reaction under emergency conditions [10]. Lindell and Prater introduced the principal behavioral variables affecting hurricane evacuation time estimates [11]. Pel et al. developed the macroscopic evacuation traffic simulation model EVAQ and analyzed the impact of trip generation, departure rates, route flow rates, road capacities, and maximum speeds on evacuation by applying EVAQ [12]. Pel et al. also reviewed travel behavior modeling in dynamic traffic simulations for evacuation [13]. However, most previous research uses only a deterministic set of parameters to represent realistic evacuee behavior, i.e. they view evacuee behavior as constant. They do not consider the impact of variation of evacuee behavior on evacuation clearance time. Therefore, new emergency evacuation management and control strategies from the perspective of evacuee behavior cannot be proposed. Evacuees’ departure time choice, destination choice and route choice behavior based on real-time traffic information are three crucial factors to affect evacuation clearance time. Different combinations of values of these three factors result in different evacuation clearance time. In this paper, we present our work on determining optimal or good enough values of these three factors simultaneously based on the orthogonal experimental design method, which can reduce evacuation clearance time significantly. Once we obtain these values, we can guide evacuees’ departure time choice, destination choice and route choice behavior to approximate these values and evacuation efficiency will improve greatly. Emergency Traffic Evacuation Control based on the Orthogonal Experimental Design Method * Lv Yisheng, Zhu Fenghua, Member, IEEE, Miao Qinghai, Ye Peijun, Chen Songhang 2012 15th International IEEE Conference on Intelligent Transportation Systems Anchorage, Alaska, USA, September 16-19, 2012 978-1-4673-3063-3/12/$31.00 ©2012 IEEE 1269