On Stochastic Optimality of Policies in First Passage Problems.
Abstract
In stochastic scheduling and optimal maintenance problems that have been considered in the literature, the optimization criterion used has often been equivalent to minimizing the expected first passage times to a set of states. A typical method used in establishing the optimality of a certain policy is the method of successive approximations on the appropriate dynamic programming functional equations. As an intermediate result, this technique often involves the optimally of the pertinent policy for all finite horizon versions of the problem. This paper characterizes stochastically optimal policies as policies that process a similar property, i.e. these are optimal in expectation for all members of a sequence of appropriately defined finite horizon problems. The authors use this characterization to establish the stochastic optimality of relevent policies for the optimal repair allocation for a series system problem and for a scheduling problem.
Document Details
- Document Type
- Technical Report
- Publication Date
- Jan 01, 1984
- Accession Number
- ADA186365
Entities
People
- Costis Melolidakis
- Michael N. Katehakis
Organizations
- Columbia University