Approach to collect leading indicators in major accident areas

Major Accidents have the potential to endanger multiple lives, the environment but also the market position of even the biggest companies in the world. In the past we have seen various major accidents in the Oil & Gas Industry, one of the most recent and best known is the Macondo Blowout a.k.a. Deepwater Horizon Explosion that killed eleven workers and which was followed by an extensive oil leak in the Gulf of Mexico. Fortunately these accidents do not happen very often, but this also makes them difficult to predict. Based on the risk of these kinds of events, Royal Dutch Shell, one of the biggest public oil & gas companies in the world and sponsor of this research, wants to know whether it is possible to develop an approach to improve early detection of errors in the prevention of major accidents. As a result, this research project proposes an approach to detect ‘leading indicators’ for major accidents. This approach is developed and tested in a ‘field’ pilot. The main research question for the project is defined as: How can lead indicators about the state of major incident prevention be collected to gain more information about the state of incident prevention of major accidents? Early in the project it was decided to take an approach that enables everyone, including non?experts to use their smart phones to report on the state of barriers at a location or site where major accidents could occur. The tool that should enable this is called MoBaC, which stands for Monitoring of Barrier Conditions, and consists of predefined questions that focus on safety?critical barriers that should prevent major accident from happening. The major accident areas that are relevant for Shell are identified as: \u95 Process Safety/Asset Integrity \u95 Well integrity \u95 Transport of people by helicopter, airplane, bus or over water \u95 Transport of oil and gas overwater or over the road Due to the time?scope of this of the project it was not possible to look into all of these areas, therefore this research focuses only on ‘Process Safety/Asset Integrity' and 'Transportation of people by bus’. To streamline the identification process for suitable questions and ensure that the questions asked do not mainly focus on ‘personal safety’ a method is derived that can help with this process. This approach is used to derive the questions for the two major accident areas mentioned above and can be summarised as follows: METHOD IDENTIFYING QUESTIONS FOR MAJOR ACCIDENT AREAS: 1. Identify the main direct causes of major incidents within the selected area 2. Per direct cause: identify barriers that should prevent that cause 3. Per barrier: identify possible visible signs that can give an indication of the state of the barrier => this results in a list of potential questions 1. Next to all of the above, subject matter experts should be asked to assess the questions and see whether they ‘miss’ anything. In some cases the remaining questions can be assessed on a scale of 1?4 according to two aspects, visibility and usability/relevance. A pilot project was executed for the area of 'Transportation of people by bus' at the Pearl GTL site in Qatar. The target group consists of 800 people who are required to take the bus every workday. The pilot project was active for more than a month. In that time the questionnaire has been entered 89 times. The results of this questionnaire with respect to the barriers that are defined for this area are scientifically inconclusive; nonetheless the local road safety manager confirmed that he thinks the results have added value for safety at his location. Later a user?experience questionnaire was sent out to all known participants of the bus?transportation questionnaire, which was entered by 16 people. It was found that the majority of these 16 people have in general a positive attitude towards the project, but due to the small amount of participants no leading conclusion can be found for the entire target group. It can be argued that every compromised barrier that is being fixed due to the use of this tool does always decrease the risk of a major accident happening at a given location; nonetheless it is not possible to provide scientific proof that this approach to collect leading indicators can have a risk?reducing effect on major accidents. Overall this project has been a very interesting learning experience and it will be continued outside the scope of this report.Systems EngineeringTechnology, Policy and Managemen