Restricted Bayes Strategies for Programs with Simple Recourse

When using linear stochastic programs one generally assumes that the joint probability distribution of the random variables involved is known. However, in practice, for example in manpower planning problems, the knowledge of this distribution is incomplete. Consequently this paper presents a linear stochastic program with partial information, where the confidence in this joint probability distribution is expressed by a parameter varying from 0 to 1. Explicit solutions are given for programs with simple recourse.