Risk analysis categories

Risk can be measured and expressed in a number of ways. CCPS s Guidelines for Chemical Process Quantitative Risk Analysis (Ref. 4) identifies three main categories of risk measure Risk Indices, Individual Risk, and Societal Risk. 

The terminology used varies considerably. Hazard identification and risk assessment are sometimes combined into a general category called hazard evaluation. Risk assessment is sometimes called hazard analysis. A risk assessment procedure that determines probabilities is frequently called probabilistic risk assessment (PRA), whereas a procedure that determines probability and consequences is called quantitative risk analysis (QRA). 

The safety data used in risk analysis can be grouped into different categories, described in the following sections. The data should be provided for raw material, intermediates, and products, as well as for reaction mixtures or wastes as they are to be handled in the process. Missing data, important in risk analysis, may be marked with a letter I, to indicate that this information is missing or as a default by a letter C, if its value is unknown but judged to be critical. 

Loss-of-Containment Causes The list in Table 23-30 indicates four basic ways in which containment can be lost. These cause categories can be used both as a checklist of considerations during the design process and as a starting point for evaluating the adequacy of safeguards as part of a process hazard and risk analysis. 

In addition to the data elements that go into the frequency and consequence analyses detailed above, there are two other categories of data that will be directly discussed. These include the selection and use of meteorological data in the consequence analysis and the population data for the evaluation of impacts. The reader may refer to Chapter 2 of the Guidelines for Chemical Transportation Risk Analysis (CCPS, 1995) for additional information on these data sources as well as any others not exphcitly discussed in this chapter. 

Detect—a safety strategy to identify potential issnes before an incident occurs. This may include periodic reviews or audits of the operation and its carriers as well as ongoing risk analysis activities to identify new opportunities to improve safety and reduce risks. Additionally under this category, detection can include techniques such as GPS tracking, which, while not preventing or mitigating a release, allows faster identification of asset location. 

Once the hazards were identified, the severity of each hazard was evaluated by considering the worst-case loss associated with the hazard. In the example, the losses are evaluated for each of three categories humans (H), mission (M), and equipment (E). Initially, potential damage to the Earth and planet surface environment was included in the hazard log. In the end, the environment component was left out of the analysis because project managers decided to replace the analysis with mandatory compliance with NASA s planetary protection standards. A risk analysis can be replaced by a customer policy on how the hazards are to be treated. A more complete example, however, for a different system would normally include environmental hazards. 

ABSTRACT This paper highlights some aspects of the many facets of electricity distribution system risk assessment - describing the different risk consequence categories which are relevant in the whole risk picture with regards to their characteristics, their type of impact and apphcable risk analysis methods. The paper illustrates that distribution system asset management constitutes of a variety of more or less conflicting objectives - and that there is no single risk assessment method which cover all the different aspects of distribution system risk. 

In the following chapters we look further into each of these risk consequence categories. The presentation is based on the authors knowledge and experience regarding the apphcation of risk assessment methods in electricity distribution first and foremost among Norwegian distribution companies. Some of the risks are well defined with respect to risk analysis methods, while others have less history of being subject to structured risk assessment. 

In (Aven 2008) three main categories of risk assessment methods are presented, as stated in Table 2. These categories are used to provide a generic grouping of the different categories of methods for risk analysis. 

Table 2. Categories of methods for risk analysis — grouping based on (Aven 2008).

Table 3. Summary of risk consequence categories, their predominant impact and risk analysis methods. 

Table 3 summarises the results of the different consequence categories, indicating the predominant attributes of the various risk consequence categories. What can be seen from the table is that there is no single method or approach which can be said to cover all aspects in one commonrisk analysis framework. It wiU rather encourage the use of many different approaches to analyse distribution system risk, depending on the type of problem. 

As explained above, self-consistent models for several outcome categories require special care in order to satisfy the constraints imposed by the laws of probability. Failure to take these constraints into account could lead to contradictory results, particularly in multiple regression models when extreme values of explanatory variables are considered. To solve this problem, a method using conditional probability identities was developed and applied this method seems to be novel in the context of risk analysis for vehicle safety. The results are models that deliver self-consistent results for every possible combination of explanatory variables as well as number of outcome categories. 

It is not possible to quantify all risks. In many cases, qualitative risk analysis is the best available. One can use a classification scheme such as that shown in Figure 34-1. Table 34-4 and Table 34-5 provide a similar scheme. An analyst assigns one of five probability categories from Table 34-4 and one of four severity categories from Table 34-5. This scheme leaves out quantities completely. Analysts need considerable judgment to apply the categories consistently. 

Risk analysis and evaluation can take place at different levels of detail, depending on the requirements at hand. In general, the available techniques can be grouped into three categories ... 

For production of medicines, relevant annexes are clustered into good manufacturing practices (main principles and several categories of products e.g. active ingredients, excipients, blood products), risk analysis, technology transfer and training materials. 

We can remark that the analysis of characteristics is done to define the class of risks. Once that is done, the process is the same in the different environments. Once the risk analysis is made, the safety requirement is to mitigate, with appropriate techniques and methodologies. Once the critical categories are identified, a strict process of software engineering is established, focused on mitigations. Like a good meal, that is cooked to perfection once the menu is established and the appropriate way to cook the dish is applied. 

The most important objective of the risk analysis is to derive a risk score based on certain criteria that objectively can identify the greatest risks to help prioritize action. The risk score analysis (Model 7.6) takes into consideration the likelihood, the exposure, and the possible consequences and equates them to a risk score under the following categories ... 

Frequency-severity risk matrices are often used in support of risk analysis. They are matrices of likelihood and consequence categories. Individual risk values of cells of the matrix are assigned using qualitative analysis. However, risk matrices are only appropriate for the prioritisation of risks that are to be further analysed or acted upon. 

In a qualitative risk analysis, probability categories must be defined. One such definition (Brauer 1990, p. 530) includes the following categories ... 

Using these probability and severity categories, a risk analysis matrix can be developed for any type of event and used to identify imacceptable risks for the operation. It can also be used to prioritize which risks wiU be addressed—where action is taken to eliminate or reduce them—and in which order. An example risk-analysis matrix is given in Figure 18.1, with high priority cells identified. The highest priority cells are located in the upper left part of the matrix, while the lowest priority cells are in the lower right corner. The approach could be used to compare the impact of many different events, such as roof faU accidents, rock burst events, and mine fires. It can also be used to combine both quantitative and quahtative risks. 

---