Agent-Based Modelling and Mental Simulation for Resilience Engineering in Air Transport

Abstract Following a discussion of recent reviews, we argue that in resilience engineering (RE) there is a need for more structured modelling approaches for analysis of resilience in sociotechnical systems that can support both qualitative and quantitative studies. In this paper we present agent-based modelling and simulation (ABMS) as an approach towards this end. An agent-based model of a sociotechnical system describes the performance and interactions of its constituent human operators and technical systems in an operational context. In support of RE it can effectively be used to analyse the capability of a sociotechnical system to deal with disturbances and performance variability. We present an RE cycle, which uses qualitative and quantitative ABMS phases for analysis of the adaptive capacity of a sociotechnical system. The focus in this paper is on the qualitative ABMS phase, including the development of a qualitative model and mental simulation using the qualitative model. The model development is supported by a set of model constructs, which represent key aspects of evolution of agents’ states and agents’ interactions. The mental simulations use reasoning on the basis of the qualitative model to structurally analyse the interactions and dynamics of the performance in the agent-based model. Results of the qualitative ABMS phase can be used to improve the resilience of operations or they may be followed by quantitative ABMS. The approach is presented in detail for aircraft runway approach operations using conventional systems and an advanced aircraft surveillance application system.