Accident steps

If sufficient experience does not exist, you should consider whether the consequence potential (Step 4) or the expected frequency of accidents (Step 5) is great. Consideration of consequence potential should include personnel exposure, public demographics, equipment density, and so forth in relation to the intrinsic hazard posed by the material of concern. In Step 5 you may perceive that the expected frequency of accidents alone is important enough to justify a QRA. However, even though your company may not have much relevant experience with the activity of interest, if the consequence potential of these accidents is not great, you may conclude that the expected frequency of the potential accidents is low enough for you to make your decisions comfortably using qualitative information alone. 

In the first step, a screening process will be applied to separate the major potential hazards these will be addressed in more detail. QRA techniques are used to evaluate the extent of the risk arising from hazards with the potential to cause major accidents, based on the prediction of the likelihood and magnitude of the event. This assessment will be based on engineering judgement and statistics of previous performance. Where necessary, risk reduction measures will be applied until the level of risk is acceptable. This of course is an emotive subject, since it implies placing a value on human life. 

In the above examples the size of the chain can be measured by considering the number of automobile collisions that result from the first accident, or the number of fission reactions which follow from the first neutron capture. When we think about the number of monomers that react as a result of a single initiation step, we are led directly to the degree of polymerization of the resulting molecule. In this way the chain mechanism and the properties of the polymer chains are directly related. 

The subsequent step is to identify the various scenarios which could cause loss of control of the hazard and result in an accident. This is perhaps the most difficult step in the procedure. Many accidents have been the result of improper characterization of the accident scenarios. For a reasonably complex chemical process, there might exist dozens, or even hundreds, of scenarios for each hazard. The essential part of the analysis is to select the scenarios which are deemed credible and worst case. 

The next part of the procedure involves risk assessment. This includes a deterrnination of the accident probabiUty and the consequence of the accident and is done for each of the scenarios identified in the previous step. The probabiUty is deterrnined using a number of statistical models generally used to represent failures. The consequence is deterrnined using mostiy fundamentally based models, called source models, to describe how material is ejected from process equipment. These source models are coupled with a suitable dispersion model and/or an explosion model to estimate the area affected and predict the damage. The consequence is thus determined. 

In Steps 2 through 5 of Figure 5 you will use subjective judgment to consider whether the situation involves major hazards, familiar processes, large consequence potential, or frequent accidents. The... 

The cost of performing the hazard identification step depends on the size of the problem and the specific techniques used. Techniques such as brainstorming, what-if analyses, or checklists tend to be less expensive than other more structured methods. Hazard and operability (HAZOP) analyses and failure modes and effects analyses (FMEAs) involve many people and tend to be more expensive. But, you can have greater confidence in the exhaustiveness of HAZOP and FMEA techniques—their rigorous approach helps ensure completeness. However, no technique can guarantee that all hazards or potential accidents have been identified. Figure 8 is an example of the hazards identified in a HAZOP study. Hazard identification can require from 10% to 25% of the total effort in a QRA study. 

Many sophisticated models and correlations have been developed for consequence analysis. Millions of dollars have been spent researching the effects of exposure to toxic materials on the health of animals the effects are extrapolated to predict effects on human health. A considerable empirical database exists on the effects of fires and explosions on structures and equipment. And large, sophisticated experiments are sometimes performed to validate computer algorithms for predicting the atmospheric dispersion of toxic materials. All of these resources can be used to help predict the consequences of accidents. But, you should only perform those consequence analysis steps needed to provide the information required for decision making. 

The frequency analysis step involves estimating the likelihood of occurrence of each of the undesired situations defined in the hazard identification step. Sometimes you can do this through direct comparison with experience or extrapolation from historical accident data. While this method may be of great assistance in determining accident frequencies, most accidents analyzed by QRA are so rare that the frequencies must be synthesized using frequency estimation methods and models. 

The frequency analysis step results in an estimate of an accident s statistically expected occurrence frequency. The estimates often take the... 

Growl and Louvar (1990) describe a three-step process which most accidents follow ... 

To be effective, the creation of a task or modification of a task through the introduction of new processes or equipment should automatically require you to develop a new or revised JHA. Jobs with many steps are usually good candidates. As stated before, you should assign each job selected a priority based on the accident potential and the severity of associated potential injuries. 

After the What-lf questions for a process step have been developed, the previously obtained Checklist is applied. Tlie team selects each Checklist item for accident potential and adds them to the What-lf list for evaluation. The checklist is reviewed for each area or step in the process. 

After developing questions, the PrHA team considers each to determine possible accident effects and list safety levels for prevention, mitigation, or containing the accident. The significance of each accident is determined and safety improvements to be recommended. This is repeated for each process step or area outside of team meetings for later team review... 

Function event trees are developed to represent the plant s response to each initiator. The function event tree is not an end product it is an intermediate step that provides a baseline of information and permits a stepwise approach to sorting out the complex relationships between potential initiating events and the response of the mitigating features. They structure plant respoases to accident conditions - possibly as time sequences. The transition labels of function event trees (usually along the top of the event tree) are analyzed to provide the probability of that function occurring or not occurring. 

Step I - Select the combinations of systems that enter the analysis. (This is equivalent to finding accident sequences in event tree analysis.)... 

Step 2 - Construct a global digraph model for each accident sequence. 

Step 3 - Partition digraph models into independent subdigraphs and find singleton and doubleton minimum cutsets of accident sequences. 

The first step-in plant-system and accident-sequence analysis is the identification of earthquake-induced initiating events. This is done by reviewing the internal analysis initiating events to identify initiating events relevant to seismic risk. For example. Table 5,1 -5 shows the initiating events that were used in the Seismic Safety Margins Research Program for a PWR plant (Smith et al., 1981)... 

Instances of chemical accidents, like to those of Section 7.1, relative to the hazards listed in Step 3 are listed. Information for these must come from various souices since there is not a single database for chemical incidents. If the plant has operating experience, this is valuable information, although the data may be sparse. 

This step takes the information from Steps 6 and 8. The frequency of an accident multiplied by the consequences is the risk. The consequences need to be in common units to get a measure of the risk. Of course, multiple consequence measures may be used and give multiple risk measures frequency of fatalities, frequency of injuries, frequency of fishkill, frequency of monetary loss. Judgment must be used to rank there relative significance. 

If Step 7 minimizes consequences, and Step 9 minimizes accident frequency, it would seem perforce the risk would be minimized and such is generally the case. However, there is a synergism when frequency and consequences are combined into risk. While the risk of a low-frequency high-consequence accident may be the same as a the risk of a high-frequency low-consequence accident,... 

The PSA (Miller, 1990, Wyss, 1990a, 1990b) consisted of three steps 1) issues important to safety were identified by "brainstorms" constructed as an accident progression event tree, 2) deterministic calculations were performed on the issues when information was not available from previous calculations or similar systems, and 3) information from step 2 was used to elicii e.vpert Judgement of the issues identified in step 1. 

Accident Sequence Quantification estimates the IE frequency. Specifically, the plant model built in the Step 2 is quantified by data from Step 3 according to Boolean algebra. Quantification may be a point-value calculation in which all parameters are delermimsiic, or as uncertain values known by their distribution function. 

Hazard identification, step one, means identification of new chemicals or other factors that may cause harmful health effects. Previously, novel hazards were usually observed in case studies or after accidents or other excessive exposures, usually in occupational environments. Today, thorough toxicity studies are required on all pesticides, food additives, and drugs. New chemicals also have to be studied for their potential toxic effects. Thus, earlier hazards were in most cases identified after they had caused harmful effects in humans. Today, most chemical products have been evaluated for their toxicity with experimental animals. Therefore, hazard identification has become a preventive procedure based on safety studies conducted before a chemical compound or product reaches the market, and before individuals are exposed to it. ... 

You may believe that the accidents could not happen at your plant because you have systems to prevent them. Many of the accidents I describe occurred on plants that had such systems, but the systems were not always followed. The accidents happened because of various management failures failure to convince people that they should follow the systems, failure to detect previous violations (by audits, spot checks, or just keeping an open eye), or deliberately turning a blind eye to avoid conflict or to get a job done quickly. The first step down the road to many a serious accident occurred when someone turned a blind eye to a missing blind (see Chapter 1). 

Such audits may therefore be useful as a method of increasing safety awareness and management commitment to safety as part of a more general attempt to reduce accidents. They should be treated as first steps and management must be prepared to do more than just carry out a safety audit. The authors of safety audits must be prepared to provide guidance on the next steps in error reduction once the problems have been identified. 

However, in the case of a root cause analysis system, a much more comprehensive evaluation of the structure of the accident is required. This is necessary to unravel the often complex chain of events and contributing causes that led to the accident occurring. A number of techniques are available to describe complex accidents. Some of these, such as STEP (Sequential Timed Event Plotting) involve the use of charting methods to track the ways in which process and human events combine to give rise to accidents. CCPS (1992d) describes many of these techniques. A case study involving a hydrocarbon leak is used to illustrate the STEP technique in Chapter 7 of this book. The STEP method and related techniques will be described in Section 6.8.3. 

---