A Fast Elimination Method for Pruning in POMDPs

Ozgen, Selim; Demirekler, Mübeccel

doi:10.1007/978-3-319-46073-4_5

Cited by 1 publication

(1 citation statement)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Pruning algorithms often aim to remove unnecessary or dominated vectors from the value function; otherwise, the number of possible vectors will increase quickly and can lead to slowing down the process of nding an optimal policy in the steps of the exact value iteration algorithm. This property is used for representing the value function by a nite set of jSj-dimensional vectors, = f 0 ; 1 ; : : : ; k g, in which the belief point b has a value of V (b) = max n 2 b: n , where \:" denotes \dot product"(for more details about this, refer to [56] or [57]). The CBSM model proposed in the present paper can also be rewritten in the vector form.…”

Section: Accelerated Vector Pruning (Avp) and Incremental Pruning (Ip) Methodsmentioning

confidence: 99%

A POMDP framework to find Optimal Policy in Sustainable Maintenance

2018

View full text Add to dashboard Cite

The increasing importance of maintenance and a cleaner environment besides the relations between them has encouraged the current authors to investigate a mathematical Markovian model for the condition-based maintenance problem while considering environmental e ects. In this paper, the problem of proposing a maintenance optimal policy for a partially observable, stochastically deteriorating system is studied in order to maximize the average pro t of the system with sustainability aspects. The modeling of this Condition-Based Sustainable Maintenance (CBSM) problem is done by mathematical methods such as Partially Observable Markov Decision Process (POMDP) and Bayesian theory. A new exact method, called accelerated vector pruning method, and other popular estimating and exact methods are applied and compared for solving the presented CBSM model, and several managerial conclusions are obtained.

show abstract

Section: Accelerated Vector Pruning (Avp) and Incremental Pruning (Ip) Methodsmentioning

confidence: 99%