An Automated Decision Support System Based on Game Theoretic Optimization for Emergency Management in Urban Environments

In the context of multiple emergencies occurring in an urban environment, it is important to perform a fair allocation and scheduling of emergency response units to each emergency, as human lives could be at risk. In this work, a multi-emergency management system based on a single step, non-cooperative, normal form game model, and a Nash equilibrium based optimization methodology is proposed. In the proposed system, each emergency event is represented as a player in the game, who is competing with other players for the allocation of resource units that are available in limited quantities within a given urban perimeter. The Nash equilibrium based methodology identifies a socially fair allocation of resources depending on various fairness criteria like the demand by each emergency event, and the criticality of the events. The fairness criterion is well modeled in the game theoretic setting, while the criticality of an event can be modeled as per the requirements of a specific emergency management system. Such a system will be useful in managing emergencies in small to medium urban settings. The proposed game theoretic methodology naturally models the emergency response and resource deployment problem in the framework of social fairness, which is pivotal in these scenarios. The Nash equilibrium solution is computed using the Terje Hansen's fixed-point algorithm. Experimental results are presented for various test cases and metrics are developed to establish the quantitative measure of fairness of the results. The proposed system can be used as a decision support tool for managing emergencies, or as a simulator for learning and training purposes.