Incident analysis behavior-based

The origin of the Domino Theory is credited to Herbert W. Heinrich, circa 1931, who worked for Travelers Insurance. Mr. Heinrich nndertook an analysis of 75,000 accident reports by companies insnred with Travelers. This resulted in the research report titled The Origins of Accidents, which concluded that 88 percent of all accidents are caused by the unsafe acts of persons, 10 percent by unsafe physical conditions, and 2 percent are Acts of God. His analysis of 50,000 accidents showed that, in the average case, an accident resulting in the occurrence of a lost-time work injury was preceded by 329 similar accidents caused by the same unsafe act or mechanical exposure, 300 of which produced no injury and 29 resulted in minor injuries. This is sometimes referred to as Heinrich s Law. Mr. Heinrich then defined the five factors in the accident sequence, which he identified as the Domino Theory. Heinrich s work is the basis for the theory of behavior-based safety, which holds that as many as 95 percent of all workplace incidents are caused by unsafe acts. See also Accident Chain Behavior-Based Safety. 

Safety training and incident analysis based on the behavioral inven-... 

Instead, keep track of the various components of an incident analysis. Monitor the number of near hit, property damage, and injury reports. Track the number of corrective actions implemented for environment, behavior, and person-based factors. Now, you have an accountability system that facilitates participation. Of course, the focus needs to be on... 

Figure 9.6 Ask ten basic questions to conduct a behavior-based incident analysis.

Figure 9.6 summarizes the main steps of a behavior-based incident analysis with a flow chart of ten basic questions to ask. Before an individual worker is targeted with a training intervention, engineering strategies are considered for task simplification. 

Specific projects or assignments may come and go, but safety teams need to work persistently on their general missions in order to achieve continuous safety improvement throughout a work culture. The membership of these teams will change periodically and team goals will vary, but the challenges of behavioral observation and fe back, incident analysis and corrective action, ergonomics analysis and intervention, and behavior-based... 

Incident Analysis Team - conducts fact-finding evaluations of near-hit reports and injuries, including behavioral, environment, and person-based factors and recommends corrective action. 

There are a large number of studies showing that when a program is instituted to reduce the frequency of unsafe behaviors, a reduction in incident rate coincides with a reduction in unsafe behaviors. Krause, Seymore, and Sloat (1999) report a meta-analysis of 73 firms that implemented behavior-based safety programs for up to five years. Reductions in accident rates from pre-implementation baselines averaged 29 percent after one year and up to 69 percent by the fifth year. Thus, there is criterion-based validity when incidents are the criterion. We are not aware of any studies that have used process safety incidents or fatalities as a criterion. 

Abstract. This paper introduces an agent-hased approach to analyze the dynamics of accidents and incidents in aviation. The approach makes use of a number of elements, including formahzation of a real world scenario, agent-based simulation of variations of the scenario, and formal verification of dynamic properties against the (empirical and simulated) scenarios. The scenario formalization part enables incident reconstruction and formal analysis of it. The simulation part enables the analyst to explore various hypothetical scenarios under different circumstances, with an emphasis on error related to human factors. The formal verification part enables the analyst to identify scenarios involving potential hazards, and to relate those hazards (via so-called interlevel relations) to inadequate behavior on the level of individual agents. The approach is illustrated by means of a case study on a runway incursion incident, and a number of advantages with respect to the current state-of-the-art are discussed. 

The process is based on the assnmption that nnsafe actions are a significant cause of accidents. The first step is to identify critical behaviors. These are behaviors that are aitical to safety. Kianse (1997) lists four ways to identify these behaviors (1) behavioral analysis of injury reports, (2) worker interviews, (3) observation of the work process, and (4) review of woik rules, job safety analysis, procedures, and so on. Krause feels that 1 is most important but is not simple. Documenting incident reports is usually not specific to behavior patterns. The analysis usually requires getting additional information from operators familiar with the actual work practices. 

The maimer in which the list of critical behaviors is developed ensures a degree of content-based validity. These are behaviors that are related to incidents or that safety professionals believe to involve risk. Krause argues that behavior is the final common path for accidents. Although this is easy to demonstrate when people faU, catch their hand in a machine, or are struck by a falling object, the connection is often much more complicated when a plane crashes or a refinery catches fire. There are undoubtedly acts and failures to act involved in each of these, but an analysis of individual behavior is a rather inefficient way to understand the root causes of these events. 

---