Sampling Risks
Any uncertainty in a project plan that potentially can be controlled, tracked or identified is termed as Risk. Risk analysis involves consideration of uncertainties.
- Sampling involves a risk that the sample will not adequately reflect the conditions in the lot.
- A quantitative probabilistic method for uncertainty analysis includes the following:
- Quantifying and assigning probabilistic distributions to the uncertain inputs.
- Sampling the distributions of uncertain parameters in an iterative pattern.
- Propagate uncertainties through any effective model.
- Predict the results in terms of probabilistic measures.
However, the results of the probabilistic analysis depend on the number of sample being selected. Sampling risk analysis demands a sample plan. Sample plan is a plan that states sample sizes and the criteria to accept or reject items. The sample size needed for a specific analysis depend on various factors such as type of model, type of distributions and so on. The general tendency is to reduce the sample as much as possible without realising the effect on decisions. For example: the mean of the output requires a number of samples that is an order of magnitude less than the variance. Hence, it is desirable to use a sampling method that can predict the output probabilistic measure accurately with the minimum number of samples. Many sampling methods exist, Quasi Monte Carlo being one among them.
In inspection procedure, the probability, under the sampling plan used, means acceptable material will be rejected or that unsatisfactory material will be accepted.