Iranian Journal of Ergonomics

Introduction: In many sensitive and sophisticated workplaces, human errors are very important. Many events occur annually around the world is evidence to confirm error due to the data processing operators. The main goal of this study was to investigate human information processing models and workload demands on human operators, then developing a regression model for this purpose.

Material and Methods: This was a case study. Task analysis of control room members, including shift supervisor, board man, and site man and user instrumentation in this study was investigated. Data collection was performed by interviewing employees using a questionnaire with two parts the first part was the model of human information processing and the second part was workload NASA–TLX and an appendix of opinions existing staffing level and finally task analysis performed using SPSS 16 – SPSS.

Results: There was a correlation between workload and stages of human information processing model such as perception )r=0.65, p<0.01), vocal response (r=0.45, p<0.01) and manual response (r=0.15, p<0.05). Some effective factors on workload were vocal response (β=0.232, p=0.033) and perception (β=0.576, p=0.00). The most important factors in the analysis of anticipation staffing demand were perception (β=2.514, p=0.008), workload (β=1.57, p=0.018) and vocal responses (β=2.428, p=0.01), respectively.

Discussion and Conclusion: Increasing concentration and attention and perception skills using psychological measures including decreasing the workload and stress and holding specialized training courses are necessary.

Background and Objectives: In recent decades, several major accidents have occurred in high-reliability industries such as petrochemical companies. Accident analysis shows that the occurrence of more than 90% of accidents in industries are due to human factor and only with technical-engineering measures and the establishment of safety rules and regulations can not be institutionalized safe behaviors in such industries. Therefore, despite a slight reduction in human presence in these industries, the potential for human error risks is still high. The aim of this study was to identify and assess human errors in a petrochemical plant using the technique for the retrospective and predictive analysis of cognitive errors (TRACEr).
Methods: The sample size was all the eight operators of control room working in four shifts. In the first step, all tasks were analyzed using the hierarchical task analysis in order to identify sub-tasks. Then, for all the subtasks, different error modes (external and internal), psychological error mechanism (PEM) and performance shaping factors (PSFs) were identified and recorded in TRACEr sheet.
Results: The analysis of TRACEr sheets indicated that of a total number of 1171 detected errors, the internal and external errors were 50.67% (n=593) and 49.33% (n=578), respectively. In this line, ̔timing/sequence̕ errors with 35.36% and 'quality/selection' errors with 30.03% were identified as the highest and lowest external error modes, respectively. In classifying the internal error modes, action errors with 31.87% and decision making with 10.73% were identified as the highest and lowest external error modes, respectively. Within PEMs, ̔distraction/preoccupation̓ (23.61%) was identified as the main causes of perception errors. The analysis of the PSFs shows that ‘Organization’ with 27.95% and ‘task complexity’ with 8.74% were two main factors affecting the task errors.
Conclusion: The current study could identify many of the errors and conditions that affect the performance of operators. Therefore, this study can be introduced as a basis for managers and stockholders of chemical industries with complexity and high risk in order to prioritize human error prevention programs.