Scenario Reoptimisation under Data Uncertainty

Many dynamic planning and management problems are typically characterised by a level of uncertainty regarding the value of data input such as supply and demand patterns. Assigning inaccurate values to them could invalidate the results of the study. Consequently, deterministic models are inadequate for the representation of these problems where the most crucial parameters are either unknown or are based on an uncertain future. In these cases, the scenario analysis technique could be an alternative approach. Scenario analysis can model many real problems in which decisions are based on an uncertain future, whose uncertainty is described by means of a set of possible future outcomes, called "scenarios". In this paper we present a scenario analysis approach to dynamic multi-period systems by integrating scenario optimisation and subsequent deterministic reoptimisation. In the scenario optimisation phase we represent data uncertainty by a robust chance optimisation model obtaining a so-called barycentric value with respect to selected decision variables. The successive reoptimisation model based on this barycentric solution allows planning a part of the risk of a wrong decision, reducing the negative consequences deriving from it.