A Multiagent Approach to Ambulance Allocation Based on Social Welfare and Local Search

During a mass casualty incident, there will be many victims who need to be driven in an ambulance to a hospital. Reasoning about which patients to assign to which hospitals can be viewed as a multiagent resource allocation issue. The approach taken in this paper is to view this as a constraint satisfaction problem that should also be sensitive to a chosen social welfare metric. Our proposed algorithm employing local search is presented and then implemented in a series of simulations which experiment with different social welfare functions. The initial state used in the search is identified as a factor in the results. Moreover, a global view of the scenario helps to decide the appropriate strategies. We conclude with a discussion of next steps for multiagent resource allocation problems during mass casualty incidents. In short, we offer a more reasoned approach for ambulance allocation that may provide guidance for effective healthcare delivery.