Scenarios have become a vital methodological approach in business as well as in public policy. When scenarios are used to guide analysis and decision-making the aim is typically robustness and in this context we argue that two main problems at scenario set level is conservatism, i.e. all scenarios are close to a perceived business-as-usual trajectory and lack of balance in the sense of arbitrarily mixing some conservative and some extreme scenarios.
The purpose of this paper is to address these shortcomings by proposing a methodology for generating sets of scenarios which are in a mathematical sense maximally diverse. The authors present a systematic methodology, Scenario Diversity Analysis (SDA), which addresses the problems of broad span vs. conservatism and imbalance. From a given set of variables with associated states, SDA generates scenario sets where the scenarios are in a quantifiable sense maximally different and therefore best span the whole set of feasible scenarios.
The usefulness of the methodology is exemplified by applying it to sets of storylines of the emissions scenarios of the Intergovernmental Panel on Climate Change. This ex-post analysis shows that the storylines were not maximally diverse and given the challenges ahead with regard to emissions reduction as well as adaptation planning, the authors argue that it is important to strive for diversity when developing scenario sets for climate change research.
The proposed methodology adds significant novel features to the field of systematic scenario generation, especially with regard to scenario diversity. The methodology also enables the combination of systematics with the distinct future logics of good intuitive logics scenarios.