Uncertainty, Risk, Reliability
Risk is defined as the combination of likelihood of occurrence and severity of the consequences of an adverse event. Uncertainty is the inverse of likelihood and is therefore a basic feature in risk analysis.
Reliable information has a degree of uncertainty associated with it which is deemed to be as low as practical in the context. Those involved need to understand the data, its provenance and shortcomings and the context of its final output.
It is important to note that 'reliable' does not imply that uncertainty has been eliminated. Predictive models are always approximations of future realities.
An option analysis is a fundamental strategy used when seeking to create a new system or make modifications to an existing system. It normally lies at the heart of engineering studies.
Typical features of an option analysis are:
- The use of a set of requirements/goals that address all relevant issues - i.e. a holistic approach to requirements.
- The creation of a set of options that have potential to satisfy the requirements.
- The gathering of reliable information about the options.
- Careful consideration of the options against the requirements leading to a decision on what, on balance, would be the appropriate way forward - i.e. the achievement of reliable outcomes.
- Those who manage the study need to be unconstrained in the quest for achieving reliable outcomes.
Figure 1 is a diagram of the process.
Strategies that may be used in the course of reliable studies include:
- Those involved in the study must have the necessary competence to be able to ensure that the outcomes are fit for purpose.
- Use the systems approach where, in addition to consideration of the parts of the system, the behaviour of the system as a whole is taken into account.
- Predictive modelling i.e. use of mathematical techniqes to estimate how the system will behave in the future.
- Quantified risk assessment i.e. use of statistical probability to assess the likelihood of events.
- Data analysis
- Sensitivity studies i.e. use of predictive modelling and probablity analysis to estimate how variation in input parameters would may affect the performance of the system.
- Piloting, prototyping
Despite all efforts to reduce the risk of planning decisions being inappropriate, residual risk should be expected leading to possible unsatisfactory outcomes. Continuous monitoring of the system should be put in place to (a) seek to identify and correct unsatisfactory features of the system and (b) to inform later planning exercises.
Independence of those carrying out the studies
It is important that the process for a reliable study is contolled by people who are free to take an independent stance i.e. are not constrained by preconceived requirements about what the outcomes should be.
It is noted that European legislation requires those proposing projects related to environmental issues to carry out studies that are 'comprehensive and reliable'..