The use of a process mining technique to characterize the work process of main control room crews: A feasibility study

• Objective criteria for assessing a work process are needed in conducting HRA.
• Process mining techniques are applied to the analysis of communication logs.
• Real-life dataset of operating crews in nuclear power plants were analyzed.
• Process discovery enables to visualize work process in the communication logs.
• Semantics of utterances can be understood through a keyword flow.

In terms of supporting HRA (Human Reliability Analysis) practitioners, one of the urgent issues is to establish a set of objective criteria for determining the proper level of PSFs (Performance Shaping Factors), which are crucial for estimating the likelihood of HEPs (Human Error Probabilities). From this concern, the feasibility study of process mining techniques to characterize the work process of MCR (Main Control Room) crews is presented in this study. Three kinds of information requirements that are essential for determining the quality of the work process are first identified, and the application of process mining techniques is then introduced to address those requirements. As a case study, we illustrate the process mining techniques with communication logs that were collected from MCR crews exposed to simulated off-normal conditions. As a result, three kinds of insightful information (i.e., a work flow, time and spatial information along with a given work flow, and the flow of keywords describing what kinds of symptoms and/or knowledge were considered by MCR crews) are soundly extracted from communication logs. Consequently, it is expected that process mining techniques are effective for identifying a set of necessary information that would helpful for assessing the quality of the work process in an objective manner.