Stochastic Sequential Decision-Making with a Random Number of Jobs

This paper addresses a class of problems in which available resources need to be optimally allocated to a random number of jobs with stochastic parameters. Optimal policies are presented for variations of the sequential stochastic assignment problem and the dynamic stochastic knapsack problem, in which the number of arriving jobs is unknown until after the final arrival, and the job parameters are assumed to be independent but not identically distributed random variables.