An integrated hazard identification method based on the hierarchical Colored Petri Net

An integrated hazard identification method based on the hierarchical Colored Petri Net
The traditional hazard analysis approaches applied to the socio-technical system cannot cover the complex organization structures, the interactions between systems and human behaviors, the interrelated factors among sub-systems and the safety culture of specific societies. This paper presents an integrated hazard identification approach named System-Theoretic Process Analysis Based on Formalization Model (BFM-STPA), which can solve the above issues. Firstly, the hierarchical control structure models of the socio-technical system are established through Colored Petri Nets (CPN) due to its strong description ability and executable ability, which may also be regarded as the graphical system specification. Secondly, the hazards can be identified according to the CPN models following a series of guide conditions. Thirdly, the comprehensive contributing factors to the hazards will be found out by tracing the former states within the reachability graph generated from CPN model. Finally, an integrated hazard log can be derived for further hazard analysis and safety-guided design. In this paper, the above method was applied in the Chinese Train Control System level 3 (CTCS-3). And the process of hazard identification for the scenario of Temporary Speed Restriction issued was elaborated in detail. Compared with the hazard log generated by Hazard and Operability (HAZOP), the hazard log generated by BFM-STPA covered not only the subsystem failures, but also the deviation of interactions among subsystems from design intent, human errors and socio-technical drawbacks related to the TSR Issued scenario of CTCS-3.

Rui Wang, Wei Zheng, Ci Liang, Tao Tang, An integrated hazard identification method based on the hierarchical Colored Petri Net, Safety Science, Volume 88, October 2016, Pages 166-179, ISSN 0925-7535, http://dx.doi.org/10.1016/j.ssci.2016.05.006.
(http://www.sciencedirect.com/science/article/pii/S0925753516300807)
Keywords: Hazard identification; Formal model; Hazard log; Chinese Train Control System