TY - JOUR
T1 - Supporting Systems of Systems Hazard Analysis Using Multi-Agent Simulation
AU - Alexander, Rob
AU - Kelly, Tim
PY - 2013/1
Y1 - 2013/1
N2 - When we create a safety-critical system, we need to perform an adequate hazard analysis. For Systems of Systems (SoS), however, hazard analysis is extremely difficult because of the complexity of SoS and the environments they inhabit. Traditional hazard analysis techniques often rely upon static models of component interaction and have difficulties exploring the effects of multiple coincident failures. We cannot, therefore, rely on them to provide adequate hazard analysis of SoS. In this paper, we present a hazard analysis technique (SimHAZAN) that uses multi-agent modelling and simulation to explore the effects of deviant node behaviour within a SoS. We define a systematic process for developing multi-agent models of SoS, starting from existing models in the MODAF architecture framework and proceeding to implemented simulation models. We then describe a process for running these simulations in an exploratory way, bounded by plausibility. This process generates extensive logs of simulated events; in order to extract the causes of accidents from these logs, we present a tool-supported analysis technique that uses machine learning and agent behaviour tracing. We evaluate the approach by comparison to explicit requirements for SoS hazard analysis, and by applying it to a case study. We believe that SimHAZAN can reveal hazards that are difficult to discover when using traditional techniques.
AB - When we create a safety-critical system, we need to perform an adequate hazard analysis. For Systems of Systems (SoS), however, hazard analysis is extremely difficult because of the complexity of SoS and the environments they inhabit. Traditional hazard analysis techniques often rely upon static models of component interaction and have difficulties exploring the effects of multiple coincident failures. We cannot, therefore, rely on them to provide adequate hazard analysis of SoS. In this paper, we present a hazard analysis technique (SimHAZAN) that uses multi-agent modelling and simulation to explore the effects of deviant node behaviour within a SoS. We define a systematic process for developing multi-agent models of SoS, starting from existing models in the MODAF architecture framework and proceeding to implemented simulation models. We then describe a process for running these simulations in an exploratory way, bounded by plausibility. This process generates extensive logs of simulated events; in order to extract the causes of accidents from these logs, we present a tool-supported analysis technique that uses machine learning and agent behaviour tracing. We evaluate the approach by comparison to explicit requirements for SoS hazard analysis, and by applying it to a case study. We believe that SimHAZAN can reveal hazards that are difficult to discover when using traditional techniques.
UR - http://www.scopus.com/inward/record.url?scp=84865623528&partnerID=8YFLogxK
U2 - 10.1016/j.ssci.2012.07.006
DO - 10.1016/j.ssci.2012.07.006
M3 - Article
SN - 0925-7535
VL - 51
SP - 302
EP - 318
JO - Safety science
JF - Safety science
IS - 1
ER -