Operations’ decision making under uncertainty: case studies on papermaking

Abstract

Operational decisions are influenced not only by the data and models available to the decision maker but also by the uncertainty in the data and in model-based predictions about the impacts of decision makers' actions. In non-linear systems the potential actions may have widely differing uncertainty associated with them. Then the decision maker must take an attitude towards risk and balance that against the expectation value of performance. In stochastic optimization, methods to deal with uncertainty have been developed. However, these methods have not been widely used in decision making about operating industrial processes. In this article, we first present a short summary of decision making under uncertainty and then suggest that the mathematical structure of stochastic optimization serves as a model for the architecture of future operational decision support systems. We demonstrate this framework by analysing four idealized operational decision cases, which are closely related to practical daily decision making tasks at paper mills. However, the explicit risk analysis introduces concepts that are new to operational decision makersoperators and engineersand thus is challenging to implement in practice.