A structured, systemic methodology to improve maritime fire safety in machinery spaces

Student thesis: Doctoral Thesis

Abstract

The primary cause of machinery space fires is cited as the release of flammable oil mist contacting unprotected hot surfaces. With this being common knowledge why do we continue to see this type of incident reoccurring on ships carrying thousands of passengers to precarious destinations? Anecdotal evidence suggests that the prime focus of maritime fire safety falls on risk mitigation, as opposed to its prevention. It also appears technological safety barriers are placed in an 'install and forget' mentality through lack of appreciation of the overall system of safety control.This belief is explained by referring to risk reduction of adding a 'layer of protection' or by supposedly higher cost-effectiveness of accident mitigation measures. The significance of this belief goes beyond the accurate understanding of the state-of-the-art of maritime safety. If it turns to be wrong, at best it can misplace precious resources aimed at safety improvements, at worst it can result in an almost inevitable drift towards a machinery space fire with neglect of the control of fire free operation.The current approach to fires in machinery spaces is hypothesised to be reactive in nature, primarily focusing on detection of already materialised hazards with little or no regard to the systemic nature of hazard occurrence. My research analyses the focus to verify the hypothesis. I then determine if an alternative method can be applied to improve fire safety.With the initial hypothesis proven correct, the societal implication could be significant. Everyday life in the 21st century involves the interactions between humans, machines and the environment, and this trend is only increasing. If safety fails to account for such factors adequately, safety related accidents may continue to occur at a consistent, if not increased, rate. Conversely with the application of the methodology developed in this research, which accounts for systemic factors using established state of the art hazard analysis techniques to a new application and in a novel manner (directly incorporating design and operation), society may see the elimination of accidents it has grown to accept as unavoidable.This thesis adopts an alternative method and process (STPA) to apply a structured approach to systemic fire hazard identification and analysis, as an input to safety barrier identification and improvement techniques (e.g. Dynamic Barrier Management [DBM], real time risk representation, etc.). Barrier identification and improvement techniques are only useful if relevant hazards and causal factors are being evaluated alongside the interactions between humans, technology, and the environment.The approach also only has success if applied into the facility lifecycle during the operational phase. The novelty of applying a structured systemic approach to achieving fire safety during HAZID with a direct link to safety auditing is therefore fundamental to the impact of this thesis.Within the research I apply this technique in a case study and audit on an operational cruise ship where it is demonstrated that the approach has unearthed otherwise dormant safety concerns. These gaps would have been difficult to discover through the application of an unstructured/ traditional approach to fire safety design and auditing. The findings from the case study provide a critical input to the improvement of the leak prevention barrier of fire safety and demonstrate the novelty and usefulness of such an approach in striving for fire free operations.Where safety barriers are required or can be improved (i.e. if the same events occur at a consistent frequency from seemingly the same causal factors), the research introduces the Systemic HAZID and Operational Risk Evaluation (SHORE) methodology which presents an easy to follow process (for those not familiar with STPA) leading to the informed selection of specific
Date of Award4 Nov 2020
Original languageEnglish
Awarding Institution
  • University Of Strathclyde
SupervisorDracos Vassalos (Supervisor) & Romanas Puisa (Supervisor)

Cite this

'