Demonstration of a Limited Scope Probabilistic Risk Assessment for Autonomous Warehouse Robots With OpenPRA

Abstract

Probabilistic Risk Assessment (PRA) is an indispensable technology to evaluate the risk, dependability, and resilience characteristics of safety-critical systems. Therefore, PRA uses widely adopted methods, such as classical event trees, fault trees, Markov chains, Bayesian networks, and their numerous combinations. To analyze challenging failure scenarios of modern, intelligent, autonomous, and highly dynamic Cyber-Physical Systems (CPS), the integration of multiple PRA methods is needed. This paper presents a PRA approach based on classical Event Tree Analysis (ETA) and Fault Tree Analysis (FTA) and provides the technical description of a new open-source software platform called OpenPRA. Besides, this paper describes a representative case study from the autonomous system domain, focusing on autonomous warehouse robots.