In this post I intended to discuss the concept of major hazards an area of health and safety that is normally the domain of specialist safety engineers or quantitative risk assessment specialists and therefore largely inaccessible to the average non-expert.
Major hazards are those threats that pose an especially significant threat to the health and safety of people. As a concept this category of hazard has been around for a relatively short period of time, originating in the 1970’s and 1980’swith the Flixborough disaster in 1974, and the Seveso plant disaster in 1976 followed by Union Carbide’s Bhopal in 1984 and Piper Alpha in 1988. These disasters highlighted the need for particular attention to be paid to such hazards due to the massive impact that they can have.
Following the Piper Alpha disaster in 1988 the Cullen Inquiry’s report resulted in the specific Offshore Installations (Safety Case) Regulations in 1992. This requirement has since become an industry standard for all high risk, major hazard installations both inside and outside the oil and gas industry.
The characteristics of a major hazard are generally recognised as being;
- Large scale technological activity
- The storage and/or use of significant amounts of energy and/or toxic chemicals
- The potential to place a substantial number of people ‘at risk’
- Special implications with regard to risk assessing, arranging emergency controls, managing hazards, or dealing with abnormal situations.
Most people believe that major hazards are restricted to the nuclear, chemical or oil and gas industries, however it is now assumed that the following all constitute major hazards;
- Nuclear power facilities
- Highly flammable, explosive or toxic chemical processing and storage facilities
- Offshore oil and gas facilities, refineries, storage facilities and petrochemical plants
- Water treatment facilities utilising bulk amounts of Chlorine
- Highly flammable material storage on a bulk scale
- Railway activities (since the King’s Cross and Clapham Junction accidents in the UK)
- All transport activities that could injure or kill large numbers of people (since the Herald of Free Enterprise disaster)
Risk assessment activities relating to major hazards tend to focus on quantitative methodologies for the purpose of delivering a formal safety case that can demonstrate the ALARP (As Low As Reasonably Practicable) principle has been achieved.
Risk assessment studies of major hazards normally begin with a credible event or failure scenario such as over pressure, fire, loss of containment, explosion or derailment in the case of a train. There are a range of tools and techniques that can then be applied to aid analysis of such hazards, for example failure mode and effects analysis (FMEA), hazard and operability studies (HAZOPS) or management oversight and risk tree (MORT). All have a quantitative focus.
The results from the analysis are then used to develop consequence risk models for the purpose of reducing hazard consequence correlations to simple and cause and effect relationships. For example a gas release can be modelled based upon the size of a container breach, wind direction and speed, volume of the container, and leak rate. The effect can then be calculated to estimate area likely to be effected, the density of gas within this area, the probable numbers of people exposed and therefore the likely effects on the population.
In relation to such major hazards there are clear benefits to using quantitative risk assessment (QRA) techniques. The first of these is the ability to demonstrate whether the risk from the hazard has been reduced to as low as reasonably practicable in the opinion of risk experts. QRA also acts as the central plank in any safety case providing as it does the design basis for improving risk controls until they meet the ALARP standard.
"Unfortunately, the dominant position of QRA and the technical risk analysts required to carry it out leaves such experts in the dangerous position of potentially propogating the myth that risk associated with a particular hazard is what the risk experts say it is!" - Waring and Glendon (1998)
As you may have read in some of my previous articles qualitative risk assessments come with their own risk. Often they are packaged up by their proponents as objective, undebatable scientific truth and therefore rarely questioned by the non-expert. However as with any other risk judgement, they remain based upon subjective human opinions. Therefore whether they are completed by experts or not such bias will affect the final analysis with the danger that it will remain unchallenged by the diverse insights that lay persons can so often bring to such debates. This is not to say that quantitative risk assessments do not have their place, only that when they are used people should be aware that they are not as neutral as their proponents often suggest.
A further danger that is that the same people who prepare the QRA for a major hazard are often the same people who make recommendations about how to apply it. This presents a conflict of opinions, reducing fresh and diverse thinking within the risk management process and should therefore be avoided.
The findings from several major disaster reviews including Piper Alpha and Kings Cross highlighted the central part played by culture and management systems. Such aspects typically require an inherently qualitative approach are difficult to quantify using technical safety techniques and require input from people with a management and social sciences background. This then avoids overly relying upon artificial, programmable and predictive models of human behaviour.
As raised by Toft (1986) and by Morgan (1986) a significant final challenge for all risk assessments but especially by quantitative risk assessors is the 'equafinity' principle. By this principle it is not possible to identify all accident and event scenarios and outcomes. This ensures that we should always be conscious that any risk assessment however well done will never be able to predict all possibilities or even to guarantee which are most likely. Something that we should all be concious of when working to control the potentially devastating impact of major hazards.
