Journal of Health and Safety at Work

---

Introduction: Safety signs provide information,related to hazards or dangers in the industry,in form of instructions. These signs are effective as long as they are designed in accordance with principles of ergonomics and design cognitive features. The purpose of this present research was to study the relationship between cognitive features of signs and ability to guess, and to develop the relevant regression model.

.

Materials and methods: This descriptive cross-sectional study was carried out on 100 employees in a petrochemical industry complex. A three part questionnaire was used to collect required data while first part of the questionnaire dealt with demographic information, second part included cognitive features of signs designand the third part proceeded on testing the ability to guess. Then, a regression model was developed to determine the relationship between cognitive features, and the ability to guess.

.

Results: Mean and standard deviation obtained for the ability to guess the total study signs were 63.73 and 4.36, respectively. The feature of “familiarity” obtained the lowest possible score (49.15). The “semantic closeness” (β=0/390) and “meaningfulness” (β=0/369) had the highest correlation with the ability to guess safety signs.

.

Conclusion: According to results of this study, use of principles of ergonomic design of signs and training are necessary to promote the ability to guess the safety signs to the minimum available standards. Therefore, it is possible to balance cognitive features especially “familiarity”, with the lowest score, and “meaningfulness” and “semantic closeness”, with the highest influential relationship with the ability to guess of signs. The developed regression model for this industry can be used to predict the ability to guess of safety signs in future studies

---

Introduction: Nowadays, human error is one of the main causes of incidents in the industry. One of the vital characteristics of modern industries is that the precise control of key parts of the process is performed by operators from central control rooms, so an error by the control room staff can be disastrous. The present study is aimed at identifying and evaluating human errors in the control room of the petrochemical industry.  
Material and Methods: This is a descriptive-analytic case study that was conducted in a control room of the petrochemical industry. In this research, firstly by using hierarchical task analysis (HTA), the tasks in the control room were identified and analyzed. Then, using the extended CREAM method, possible human errors were identified, their cognitive category was determined, and their probabilities were calculated using a new approach based on BN.
Results: The results of the study showed that the most prevalent control modes for the Boardman and the senior board man were strategic and scrambled modes with error probabilities of 0.136 and 0.171, respectively.
Conclusion: According to the results obtained in the modeling section, BN can be proposed as an approach with high processing accuracy and also high accuracy in modeling human errors and problems with high input parameters affecting the output parameter.