An AI Planning Approach to Emergency Material Scheduling Using Numerical PDDL

Abstract

Emergency material planning and scheduling aims to schedule emergency material to destination efficiently to reduce property losses and personal casualties caused by material shortage. Existing researches usually rely on optimization of transportation routes. They ignore the rich scenarios in which human and material resources participate in scheduling simultaneously. Moreover, sudden emergencies may involve the supply of multiple material demand points. To address these issues, we propose an AI planning approach to model emergency material scheduling domain and construct a standard planning task using PDDL. Then a state-of-the-art planner is employed to solve the generated planning task. Experimental results show that the proposed approach can fit the actual situation of emergencies and give high-quality instruction to guide emergency material planning and scheduling.