Research on the Analysis Method of Production Safety Accidents Based on Accident Event Causal Association Diagram

A production safety accident is a series of events that begin with the loss of control of a system and end with the system being controllable, which causes harm (damage, injury/death, and economic losses) or has the potential to cause harm. Due to the non-repetitiveness of accident processes, analyzing past accidents is an important approach to uncover human errors. Accident causation analysis must be based on a clear understanding of the accident-related events, which not only refers to the system loss of control event that triggered the accident but also includes the associated events that led to its occurrence, as well as a series of emergency and subsequent events that occurred after the accident. Currently, the analysis of accident causes is based on accident causation models. However, analysts lack a clear delineation of the series of events during the accident process and their nature and relationships. Besides, there is a subjective element in determining the basic events of the accident, resulting in varying results when using accident causation models for analysis. As a result, these analyses cannot comprehensively and objectively reflect the key reasons that led to the accident. Thus, this study introduces a novel method for production safety accident analysis-the Accident Event Causal Association Diagram (AECAD), which aims to segment the accident based on accident investigation reports and clarify the causal relationships between all events related to the accident, to provide accurate and clear accident data for the subsequent application of accident causation models. To validate the comprehensiveness, effectiveness, and convenience of the AECAD, based on the Accident Event Causal Association Diagram, this study conducted an accident causation analysis of a material handling accident that occurred in a mine in the USA using the Human Factors Analysis and Classification System model (HFACS).