MART: Major Accident Risk Tree for Investigation of Process Industries Accidents

摘要

The investigation of major accidents is required by law, and brings important information to repair the safety system and prevent further such incidents as well. A number of methods to investigate accidents at work are presented in the literature, but much less of such methods are available for process accidents causing the release of hazardous chemicals. Additionally, the teams carrying out investigations of accidents have a serious problem in establishing facts, because the systems after catastrophic accidents are often completely destroyed, there is a large number of process data, and sometimes the most important witnesses are dead. So the best strategy for overcoming these difficulties is to introduce to the investigation process knowledge and information resulting from previous experience that is available in the databases of the incident and also in the process safety standards. That is the main premise of undertaking research in this field. The paper proposes a logical diagram of major-accident investigation process, called MART model that uses the existing MORT model which is used for accidents at work. MORT model was completed with the basic principles of the formation of an accident scenario, according to the methodology of layer of protection analysis (LOPA), knowledge of the causes of serious process failures and knowledge of safety measures used in process industries. MART diagram can be used to support the investigation process and to make recommendations for improving safety performance. The Imperial Sugar-2008 case study illustrates the application of the method.