Timed pattern-based analysis of collaboration failures in system-of-systems

Abstract

A system-of-systems (SoS) tries to achieve prominent goals, such as increasing road capacity in platooning that groups driving vehicles in proximity, through interactions between constituent systems (CSs). However, during the collaboration of CSs, unintended interference in interactions causes collaboration failures that may lead to catastrophic damage, particularly for the safety–critical SoS. It is necessary to analyze the failure-inducing interactions (FIIs) during the collaboration and resolve the root causes of failures. Existing studies have utilized pattern-mining techniques to analyze system failures from logs. However, they have three limitations when applied to collaboration failures: (1) information loss caused by the limited capabilities of handling interaction logs; (2) limitations in identifying multiple failure patterns in a log; (3) absence of an end-to-end solution mapping patterns to faults. To overcome these limitations, we propose an FII pattern mining algorithm covering the main features of SoS interaction logs, an overlapping clustering technique for multiple pattern mining, and a pattern-based fault localization method. In experiments conducted on platooning and mass casualty incident-response SoS, the proposed approach exhibited the highest pattern mining and clustering accuracy and achieved feasible localization performance compared with existing methods. The findings of this study can facilitate the accurate analysis of collaboration failures.