Decision making assessment in industrial process simulation

Human factors play a major role in process industry to increase safety. The problem of assessing human performance and identify most critical human factors is very complex in a real plant, while simulation in a virtual reality (VR) context provides freedom to test and train for risky situations. On the other hand not all aspects of human factors can be easily assessed in a simulated VR environment. We can distinguish operative skills from more cognitive ones, like, e.g., communication, situation understanding, operation planning, and problem solving. While operative skills could be easily trained and tested also in a real context, communication, command, decision-making and problem solving are more difficult and can be better assessed in a VR context, by controlling the execution of simulation experiments. In this paper we present an approach to assess cognitive human factors to improve safety. The approach proposes a way to automatically detect critical actions and events necessary to assess and measure performances of command and control roles, which can also be extended to measure and assess operative roles performances.