A root cause analysis approach for maritime accident problem investigation

As more than ninety percent of global trade is carried by sea, the safety of international shipping vessels is important in terms of the health of the global economy. Maritime transport involves many danger due to the unique nature of ship operations. Mistakes to be made in such a hazardous environment may cause maritime accidents which lead to devastating consequences such as pollution and loss of life besides material and moral damages to the shipping company. For the development of maritime safety and the more effective protection of marine environment, the accidents occurred on board should be investigated systematically and effectively. Investigation of marine accidents and incidents provides an opportunity to understand the root causes and to make the necessary arrangements by identifying the risk mitigating strategies to prevent recurrence of such events. Although the importance of reporting and examination of safety-related incidents is emphasized in various international laws, conventions, codes and recommendations, marine accidents still occur. In accordance with the IMO requisition, all incidents and accidents should be fully reported in a timely manner and analyzed in order to prevent their recurrence. All this process is performed according to the Company's instructions. However, the analysis techniques used by shipping companies are different from company to another and there is no standard root cause analysis technique for ship accidents in maritime field. While some shipping companies use too simple analysis techniques such as five whys technique which is not practical for comprehensive investigation of a complex ship accident, some companies use various software which have different deficiencies. Carrying out an appropriate accident analysis is of great importance in terms of correct understanding of the root causes of the accident and production of the corrective/preventive actions. During the last decades, a number of accident investigation methods have been developed and presented in the literature. Unfortunately, the use of existing root cause analysis techniques is not sufficient in some cases due to the unique nature of ship operations. Many shipping companies do not use these methods because there is no responsible person experienced in these methods who can determine which technique ought to be used for a specific ship accident. Due to the absence of a standard root cause analysis technique in maritime industry, all shipping companies are evaluating the accidents through the forms they developed. This situation causes a long term delay in determination of the steps to be taken to handle ship accidents in a professional manner and to ensure maritime safety. Although the existence of accident evaluation techniques such as Human Factor Analysis & Classification System (HFACS) and Formal Safety Assessment (FSA) is a positive step to ensure safety, these techniques can not meet the needs exactly for reasons such as to be not useful in case of lack of data and to be insufficient to analyse organizational factors. Even the lack of standard terms for root causes determined by accidents analysis creates problems in learning from accidents and in the monitoring phase of lessons learned. Accident/incident investigation reports aiming to determine the root causes of accidents can be given as an example to illustrate this situation. When we compare accident reports prepared by different companies, we can see the difference in the depth of the analysis and changes in the terminology used. At the same time, a quantitative evaluation result is not given. It is clear that there is a need for a root cause analysis technique in which a standard terminology unique to maritime accidents is used, quantitative results are obtained, risk mitigating strategies are recommended, responsibility in fulfilling the corrective/preventive actions is shared. This study aims to fill the gaps in this regard. In the thesis, the existing root cause analysis techniques a number of which are used in maritime accident investigation are examined. Standard particulars used for comparison of these techniques are reviewed and the advantages and limitations of these analysis techniques are determined. The characteristics that a root cause analysis method should have in order to be applied in ship accidents are discussed and all these techniques are compared according to the described characteristics. In order to develop the proposed technique, it was observed more convenient way to take advantage of the professional work carried out in the aviation sector in order to develop a standard taxonomy for ship accidents. For this purpose, the steps for developing a standard taxonomy in Aviation Safety Action Program (ASAP) which is used to report safety related issues voluntarily were followed. One of the most important contribution of this study on academic literature is that standardization is achieved in the expression of ship accident root causes. Existing ship accident investigation reports of maritime accident investigation boards from different countries, MAIB database, Turkish tanker shipping company inspection database, common taxonomies used in soft wares and a U.K. shipping company database are reviewed during taxonomy development phase. The developed taxonomy is validated by performing a study by using randomly selected ship accident reports from different countries. The proposed taxonomy provided 94.1 % validity. The proposed technique consists of four steps. In the first step, evaluation criteria to be used in accident analysis are discussed in strengths, weaknesses, opportunities and threats which are the basic building blocks of SWOT analysis. Quantitative fuzzy SWOT approach is considered as appropriate in the development of the technique. Predefined strength, weakness, threat and opportunity pools are developed. In the second step, corrective/preventive action strategies are predefined. These strategies are the possible SO, ST, WO, WT strategies. In the third step, weighting operations are performed. At this stage, Chang's extended analysis is used due to having easy steps and its widespread use in the literature. In the existing literature, many studies applied the F-AHP method omit or does not report consistency control. Necessary attention is given to consistency of decision matrices in the present study. In the fourth and last step of the technique, The percentage of the responsibility of the stakeholders to implement corrective/preventive actions could be expressed quantitatively. Thus, a cooperation to prevent recurrence of ship accident will be achieved. Using the outcomes of the study, a base of a new software which will be used to perform accident investigation by shipping companies is developed. Maritime sector practice is performed and it is seen that it can easily respond to the needs of maritime sector. Examination of the root causes of ship accidents within a certain hierarchical structure makes the proposed study more convenient. It gives quantitative results and helps define the issues to be given priority. Since the consistency analysis is performed for each decision matrix, it helps ensure that the decisions taken are fast and consistent. As the analysts use predefined pools, they will not be exhausted. Further, the proposed technique contributes to increase the awareness of all stakeholders on the subject. Avoiding from recurrence of the accidents assist the transition from reactive approach to proactive approach. The proposed technique is not specialized for a particular ship type or ship operation. It has a structure which can easily be kept up to date by new standards and regulations. This research will serve as an example to fit existing maritime accident reports to a certain standard. The whole work performed in this thesis will contribute to the development of a safety culture in maritime sector.