Review of Theories and Accident Causation Models: Understanding of Human-Context Dyad Toward the Use in Modern Complex Systems

Many accident causation models/theories have been dominating the human factors literature from a range of viewpoints and in a variety of different industrial contexts. However, many of the theories/models have limited applications with respect to capturing the underlying accident causations in modern complex systems. As a result, we seem to be moving toward new incident potentials, particularly in high-risk industries. As observed, major accidents keep occurring that have similar systemic causes in different contexts. This creates serious doubts that the existing analysis methods are capable of discovering the underlying causality in new complex settings, and/or how do we transfer what we learn from past to new contexts. Many of the approaches to safety focus on enforcing 'defenses,' i.e., physical, human, and procedural barriers. This perspective has limited view of accident causality, as it ignores today's changing factors in emerging complex work systems and their interactions that generate various potentially unintended situations and shape the work behavior. Understanding and addressing these causal factors is therefore necessary to develop effective accident prevention strategies. Thus, this paper reviews the popular theories/models, by focusing on the developments in modern industrial environments toward more complex systems and interactive or collaborative operations. Accident models founded on basic systems theory concepts, which endeavor to capture the underlying accident causality and complexity of modern socio-technical systems from a broad systemic view, are analyzed and new insights for future accident analysis efforts are identified.