Inverse reinforcement learning to assess safety of a workplace under an active shooter incident

Abstract

Active shooter incidents are posing an increasing threat to public safety. Given the majority of the past incidents took place in built environments (e.g., educational, commercial buildings), there is an urgent need for a method to assess the safety of buildings under an active shooter situation. This study aims to bridge this knowledge gap by developing a learning technique that can be used to model the behavior of the shooter and the trapped civilians under an active shooter incident. Understanding how the civilians respond to different simulated environments, a number of actions can be undertaken to bolster the safety measures of a given facility. This study provides a customized decision-making tool that adopts a tailored maximum entropy inverse reinforcement learning algorithm and utilizes some safety measurement metrics, such as the percentage of civilians who can hide/exit in/from the system, to assess a workplace’s safety under an active shooter incident. For instance, our results demonstrate how different building configurations (e.g., location and number of entrances/exits, hiding places) play a significant role in the safety of civilians under an active shooter situation. The results further demonstrate that the shooter’s prior shooting experiences, the type of firearm carried, and the timing of the incident are some of the important factors that may pose serious security concerns to the civilians under an active shooter incident.