A Hierarchical Framework to Mitigate the Risk of Hazardous Material Transportation

Abstract

An integrated traffic control policy for hazardous materials (hazmat) transportation is devised based on dual toll pricing (DTP) and network design (ND) policies within a two stage simulation-based optimization framework to enhance public safety in a highway. This integrated policy is simultaneously able to restrict hazmat traffic from freeways in densely populated areas via ND and control both regular and hazmat vehicles in tollways via DTP. Moreover, the mixed integer progrmming is employed to find the optimum integrated policy, and linear-relaxation technique based on Kaush-Kahn-Tucker (KKT) conditions is adopted to reduce the search space of the optimization process. In the proposed framework, all suggested policies are evaluated by an agent-based simulation model, which is able to interpret complex interrelationship between road conditions and vehicles. In addition, supervised learning (i.e., random forest algorithm) has been implemented within the agant-based simulation model to predict the risk taking behavior of the drivers considering the drivers' race and gender and transportation network characterstics such as levelness and curviness of the roads. The proposed framework has been demonstrated with a real traffic data of San Antonio, Texas under AnyLogic® platform. The experimental results reveal that the proposed framework is able to efficiently find the optimum integrated policy which effectively reduces the risk of hazmat transportation in a highway.