Decisions under preliminary information: rush and be wrong or wait and be late?
Many decisions in Operations management involve a choice between early action at low cost but with little information and delayed action with full information, but at a higher cost. For example, in managing a supply-chain, one can postpone some orders until more demand information becomes available. This decision, also known as Quick Response, has been analyzed as a multi-stage newsvendor model with Bayesian updating. A similar problem exists in the context of product development, where often design decisions have to be made despite residual uncertainty about market requirements or inputs from other information providing activities. In this context, however, the uncertain information is not a continuous parameter (such as demand), but a discrete set of non-comparable (design) alternatives. Thus, existing models from the supply-chain literature are not applicable. In this article, the authors develop a generalized model of decision making based on preliminary information and present applications in the fields of product design and supply chain management.
Copyright INSEAD. All rights reserved