Visualising risk in generating capacity adequacy studies using clustering and prototypes

Generating capacity adequacy studies play a significant role in long term capacity planning. Risks of capacity deficits are usually reported in the form of one or more average quantities, which cannot fully convey the nature of the risks being faced. Chronological Monte Carlo simulations may be used to construct comprehensive multi-dimensional risk profiles, but such profiles tend to be difficult to interpret. This paper proposes the use of a clustering method to partition the risk profile into clusters of similar outcomes with associated probabilities. The results are presented in accessible tabular form, and prototypical scenarios can be analysed in detail to provide further insight.