A Bayesian Belief Network Approach for Integrating Human and Organisational Factors in Risk Analysis: A Case Study for the Maritime Industry

The paper presents an innovative approach to integrate human and organisational factors (HOF) into risk analysis. The approach has been developed and applied to a case study in the maritime industry, but it can also be utilised in other sectors. A Bayesian Belief Network (BBN) has been developed to model the maritime transport system (MTS), by taking into account its different actors (i.e., ship-owner, shipyard, port and regulator) and their mutual influences. The last ones have been modelled by means of a set of variables whose combinations express the relevant functions performed by each actor. The Bayesian Belief Network model of the maritime transport system has been used in a case study for the quantification of human and organisational factors in the risk analysis carried out at the preliminary design stage of high speed craft (HSC). The analysis has focused on the frequency assessment of collision, grounding, contact and striking events by means of a Fault Tree Analysis (FTA). The approach allowed identification of probabilistic correlations between the basic events and the Bayesian Belief Network model of the maritime transport system, representing the operational and organisational scenario. The linkage can be exploited in different ways, especially to support identification and evaluation of risk control options also at the organisational level. Conditional probabilities for the Bayesian Belief Network have been estimated by means of experts’ judgments, collected from an international panel of different European countries. Finally, a sensitivity analysis was carried out over the model to identify configurations of the maritime transport system leading to a significant reduction of accident probability during the operation of the high speed craft. more than 15,000 accidents in a time span of ten years. Lloyds’ statistics show that an uncorrected course and an excessive speed with respect to the traffic in the sea zone are responsible for about 50% of all the maritime accidents, particularly groundings. Moreover, 70-80% of the accidents are due to human mistakes or other events attributed to the human behaviour. MisjudgementPilot 34% MisjudgementMaster 11% Other human factors 13% Misunderstanding 9% Inattention-Pilot 13% InattentionOfficer of the watch (OOW) 10% LackCommunic. 10% Figure 2. Typology of human factors causing an accident. While technical solutions will continue to play an important role, there is widespread agreement that the key means of tackling the human element contribution to accidents will be via safety management, including inspection and training. Starting with a deeper understanding of the role of the human element in the safety performance of maritime transport, a new issue is emerging; indeed, the official report concerning the Zeebrugge incident (capsizing of a passenger ship) (Rasmussen, 1997) already pointed out that it was not due to a coincidence of independent technical failures and human errors, but a systematic change in the organizational behaviour of operators under the influence of economic pressure in a strongly competitive environment. Thus, a systematic safety analysis of the MTS needs to be enlarged to include interactions and effects of decisions taken by various actors of the MTS, and workplace and context conditions, including the economic pressure affecting the maritime sector. Various parties (operators, shipyards, regulators and government) in their respective working contexts are very often involved in a sequence of events leading to an accident; this is the most critical issue in developing an effective risk or accident analysis. The error of the operator onboard a ship is only the final act of a long and complex chain of organisational and systemic errors (i.e. the so-called latent failures). Rasmussen (1997) highlighted the conflicting interactions between parties in MTS, evidenced by his accident analysis of oil tankers and ferryboats (Shell 1992; Estonia 1995; Stenstrom 1995). The need for a systemic approach to analyse the MTS safety is therefore clear, not only focused on mistakes and violations of the operators, but also aiming to find, if they exist, the causes at the various levels of the socio-technical system which competes for determining the accidents. The IMO (International Maritime Organisation) provides a rational and systematic approach for assessing risk in shipping activity: the Regulatory Influence Diagrams (RIDs) (Guidelines for the application of Formal Safety Assessment (FSA) for use in the IMO rule-making process, 2002). The RID’s methodology is aimed at better understanding the underlying influencing factors and their effect upon performance and thus upon possible accidents (Fig. 3). Then RID analysis facilitates holistic modelling of all the influences upon an accidental event. An Influence Diagram identifies the factors, especially the underlying ones, that have a strong degrading effect upon performance and that are prime candidates for improvement: once identified, they can be analysed in more depth to understand why performance is poor and how it can be improved. Also Bayesian Belief Network (BBN; Pedrali et al., 2004) has been used for the purpose of integrating consideration of human and hardware failures and reflecting the hierarchical nature of influence domains. Thus the BBN represents an Influence Diagram in which the effects of such factors are represented in terms of conditional probability. Figure 3. Nested system of influences in Regulatory Influence Diagram (Maritime and Coastguard Agency (MCA) Press Releases) Moreover, from a risk reduction standpoint, the European Commission (EC) funded a project, called Safety at Speed (S@S; GRD1-2000-25563), to develop a Functional Model (FM) of the MTS. The FM helps in identifying the critical interactions among actors that can trigger unwanted events and the parameters we can act upon to improve the situation. The objective of the project was to develop a formal methodology for the safety design of High Speed Craft (HSC) using state-of-the-art techniques and tools. At the same time, S@S wanted to promote a safety-culture approach for integrating safety effectively in the process of ship design. Specifically, the research project developed a set of Fault Trees (FTs) representing different hazard scenarios (colli-