An enhanced CREAM with stakeholder-graded protocols for tanker shipping safety application

Safety is always a critical focus of the shipping industry, and it is well recognised that human factor is one of the major causes of accidents and breakdowns in shipping sector. In this study, a quantitative cognitive reliability and error analysis method (CREAM) is developed that provides point estimates of the human error probability (HEP) in tanker operational safety. In effect, the quantitative CREAM incorporates weightage of common performance conditions (CPCs) that are graded by the stakeholders (experts and seafarers). As different from the nine standard CPCs in the original CREAM, eight customised CPCs are proposed to better capture the essential aspects of the work situations and conditions for on-board tankers. The weighting of the CPCs is calculated by fuzzy analytical hierarchy process (FAHP), which provides reliable and practical results, while Markov method is applied to calculate human reliability and the mean time to human error. An exemplified application using five seafarers' reliability analysis generated quantified results of human performance failure consistent with that produced by the original CREAM. The quantitative CREAM developed in this research is useful for evaluating the reliability of on-board crew members, which contributes to a safety and more reliable shipping industry.