A greedy heuristic algorithm to solve a VRP-based model for planning and coordinating multiple resources in emergency response to bushfires

Abstract

Uncoordinated responses are problematic in various operations, especially in crucial circumstances such as bushfires, where they are associated with decreased efficiency, effectiveness, and safety. This paper presents a new VRP (Vehicle Routing Problem) model for routing, scheduling, and coordinating bushfire-related resources including Ground Resources (GR) and Aerial Resources (AR) in the suppression phase of a bushfire event. The coordination must be in a way that ARs' operations are allowed prior to GRs' operations, otherwise, aerial interference is unhelpful. The problem is NP-hard (Non-deterministic Polynomial-time) and cannot be solved in a Polynomial-time using exact methods, but due to the crucial circumstances of bushfires, it should be solved in a reasonable time. Therefore, we proposed a greedy heuristic algorithm to solve it. Dividing bushfires' area into a set of fire sites, we solved instances with different numbers of fire sites using both CPLEX and the greedy algorithm and compared the results. CPLEX fails to solve the instances having more than three fire sites, but the greedy algorithm solves the largest instance having 9 fire sites in less than 1 minute. The ignorable Relative Percent Difference (RPD) of the greedy algorithm in the first five instances indicates that our proposed algorithm is reliable.