Safety Modelling

Safety synthesis of an engineering system is usually conducted by aggregating safety assessments for its sub-systems, components and failure modes. A safety assessment framework may constitute a hierarchical structure with failure modes at the bottom level (Wang et al. (1996)). A failure mode could be described in several ways, for example in terms of failure likelihood, consequence severity and failure consequence probability using linguistic variables. This is a natural and sensible way for capturing ambiguity and uncertainty inherent in safety assessment. 

Fuzzy sets are well suited to characterizing linguistic variables by fuzzy memberships to the defined categories for a particular situation. Failure likelihood, consequence severity and failure consequence probability could all be characterised using the same set of categories but different membership functions. In this way, the safety associated with a failure mode may also be modelled using fiizzy sets. 

For example, the fiizzy safety description (5) associated with a failure mode can be defined as the following product of the zzy sets of the related failure likelihood (L), consequence severity (O and failure consequence probability ( ) (Wang et al. (1995,1996))  

Similar fuzzy sets could be generated for describing the safety of other failure modes, which could be aggregated using conventional fuzzy operations to generate safety descriptions for the components, the subsystems and the whole system of the assessment hierarchy. However, this process may lead to information loss. 

There are other ways to describe safety. The simplest approach would be to use a scale for scoring the safety of a failure mode. While this may be easy for safety aggregation to produce an average indicator about system safety, it could not capture uncertainty inherent in safety assessment and thereby the credibility of such assessment may become questionable. Unfortunately, several well known multiple criteria decision analysis methods, which could be used for safety synthesis, can only be implemented using certain types of scores. This will be discussed in detail in the next section. 

---

Lagasse, R. (2002). Anesthesia safety Model or myth. Anesthesiology, 97, 1609-1617. 

Occupational safety management concentrates on the safety of individual workers by promoting their safety-mindedness the prevailing view of human error is that of the traditional safety model where safety control is handled by motivation, and punishment, (for lack of attention). 

Figure 1.1 Swiss Cheese Process Safety Model (CCPS, 2007b)...

Quintana, R., Carnet, M., and Deliwala, B. "Application of a Predictive Safety Model in a Combustion Testing Environment." Safety Science 38 (2001) 183-209. 

Figure 4.1 also illustrates the additive nature of a job s safety risks. That is, added safety risks associated with each factor cumulatively increase the job s overall safety risk. The additive nature of risk is well captured by Reason s (1990) Swiss cheese safety model. Each factor represents a risk layer, and the probability of an accident increases as the number of risk layers increases. Where possible a job which a new employee is entering should be striped of as many risk layers as possible. As will be discussed below, more senior employees, who are experienced, are likely to be better able to cope with job risks which are difficult or impossible to remove. 

This paper aimed to find the regularity of the direct cause of gas explosion, through the study of analysis of the generic unsafe act in the coalmine major gas explosion accidents based on the behavior based safety model, the following conclusions can be drawn ... 

Safety approaches based on systems theory consider accidents as arising from the interactions among system components and usually do not specify single causal variables or factors [112]. Whereas industrial (occupational) safety models and event chain models focus on unsafe acts or conditions, classic system safety models instead look at what went wrong with the system s operation or organization to allow the accident to take place. 

This study goes beyond much of the earlier research and— following the approach of Hunt and Habeck (1993) and Hunt et al. (1993)—seeks to estimate the role of HRM practices in the determination of workers compensation costs in a multivariate framework. It uses a workplace safety model that incorporates a wider variety of HRM practices than has been previously employed. In particular, it analyzes the impact of the three important dimensions of HRM practices on safety employee participation in decision making, employee participation in financial returns, and the firm s management safety culture. In addition, this is the first study to consider file effect of each of these factors on claim frequency and claim severity, and to ask whether any observed change is file result of changes in technical efficiency or moral hazard (principal-agent) incentives. 

As was stated previously, it is common for advocates of the worker-focused behavioral-safety model to state that 80-95% of accidents are principally caused by unsafe acts of workers, and, therefore, the proper action is to develop worker-focused solutions. That creates the impression in the minds of managements and many safety practitioners that the workers are the problem, that all risk situations can be resolved by worker observation techniques and positive reinforcement, and that the antecedents that derive from the design of the work methods and the workplace can be ignored. 

Markov models are a reliability and safety modeling technique that uses state diagrams. These diagrams have only two simple s)rmbols (see Figure 5-17) a circle representing a working or a failed system state and a transition arc representing a movement between states caused by a failure or a repair. Solution techniques for Markov models can directly calculate many different metrics compared to other reliability and safety evaluation techniques (Ref. 9). 

A Markov system is defined as a "memory-less" system where the probability of moving from one state to another is dependent only upon the current state and not past history of getting to the state. This is the primary characteristic of a Markov model. Markov models are well suited to problems where a state naturally indicates the situation of interest. In some models (characteristic of reliability and safety models) a variable follows a sequence of states. These problems are called Markov chains. 

Prescott, D.R, Andrews, J.D. 2005. Aircraft Safety Modelling for Time-Limited Dispatch. Proceedings of the Annual Reliability and Maintainability Symposium 139-145. 

Characteristics of Complex Industrial Systems and Processes. WP6 - Task 6.1. Methods-algorithms for evaluating unknown parameters of system operation processes, reliability and safety models - a preliminary study. WP6 -TaskO.l.l 16.03.2009. Poland-Singapore Joint Research Project, 2007-2010. 

Figure 2. Systemic concept of flight safety model according to Hawkins.

The purpose of this research is to construct probabilistic safety models for a typical loop-type FBR plant so that an overall safety assessment can be performed. It is expected that (1) a systematic evaluation on the plant safety is conducted based on the quantitative analysis, (2) the insights on measures to enhance system reliability and safety are provided, (3) the operation and maintenance procedures are established based on a risk-based consideration, and (4) useful information is given to the development of basic policy on safety design and evaluation of a large LMFBR. 

Due to the aforementioned discrepancy in data availability (especially relevant to translation of toxic effect) and the fact that many clinical endpoints are multi-mechanistic, it is important to stress that each computational step should be well defined and model small steps, for example, a traditional quantitative structure-activity relationship (QSAR) approach based on chemical structure is probably relevant to distinguish hERG binders from nonbinders, but not relevant to model a small set of diverse compounds associated with a complex endpoint such as drug induced liver injury (DILI). A second important factor to consider when construchng in silico safety models is the intended use of the model, and the potential cost associated with false positives versus false negatives from the model. For instance, there is zero... 

---