Design for postponement: A comprehensive characterization of its benefits under unknown demand distributions

Recent papers have developed analytical models to explain and quantify the benefits of delayed differentiation and quick response programs. These models assume that while demands in each period are random, they are independent across time and their distribution is perfectly known, i.e., sales forecasts do not need to be updated as time progresses. In this paper, we characterize these benefits in more general settings, where parameters of the demand distributions fail to be known with accuracy or where consecutive demands are correlated. Here it is necessary to revise estimates of the parameters of the demand distributions on the basis of observed demand data. We analyze these systems in a Bayesian framework, assuming that our initial information about the parameters of the demand distributions is characterized via prior distributions. We also characterize the structure of close-to-optimal ordering rules in these systems, for a variety of types of order cost functions.