APPENDICES
APPENDIX 1
THE IMO/ILO PROCESS FOR INVESTIGATING HUMAN FACTORS
The following is a process that provides a step-by-step systematic approach for use in the investigation of human factors. The process is an integration and adaptation of a number of human factor frameworks-SHEL (Hawkins, 1987) and Reason's (1990) Accident Causation and generic error-modelling system (GEMS) frameworks, as well as Rasmussen's Taxonomy of Error (1987).
The process can be applied to both types of occurrences, i.e., accidents and incidents. The process consists of the following steps:
.1 collect occurrence data;
.2 determine occurrence sequence;
.3 identify unsafe acts/decisions and unsafe conditions;
and then for each unsafe act/decision,
.4 identify the error type or violation;
.5 identify underlying factors; and
.6 identify potential safety problems and develop safety actions.
Steps 3 to 5 are useful to the investigation because they facilitate the identification of latent unsafe conditions. Step 6, the identification of potential safety problems, is based extensively on what factors were identified as underlying factors. At times, an unsafe condition may be a result of a natural occurrence; in that case, the investigator may jump from step 3 to step 6. At other times, an unsafe act or decision may result from an unsafe condition which itself was established by a fallible decision; in such a case, the investigator should proceed through steps 3 to 6.
Step 1-Collect occurrence data
The first step in the human factors investigation process is the collection of work-related information regarding the personnel, tasks, equipment, and environmental conditions involved in the occurrence. A systematic approach to this step is crucial to ensure that a comprehensive analysis is possible and that the logistical requirements of collecting, organizing and maintaining a relevant occurrence related database are met.
For complex systems, where there are numerous interactions between the component elements, there is constant danger that critical information will be overlooked or lost during an investigation.
Use of the SHEL model as an organizational tool for the investigator's workplace data collection helps avoid downstream problems because:
.1 it takes into consideration all the important work system elements;
