Maintenance policies for systems with condition monitoring and obvious failures

Maillart, Lisa M.

doi:10.1080/074081791009059

Cited by 91 publications

(85 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…As an initial step in this direction, future work could consider the same framework considered here, but with obvious failures and investigate how the necessary conditions can be strengthened to establish the same structured policy results along ordered sample paths. For this problem, in which observing no failure in a particular period yields imperfect information regarding the underlying deterioration state, structural results have been examined along straight lines of information states (Ohnishi [4], Maillart [3]), but not sample paths.…”

Section: Resultsmentioning

confidence: 99%

Structured maintenance policies on interior sample paths

Maillart

Zheltova

2007

Naval Research Logistics

Self Cite

View full text Add to dashboard Cite

Abstract:We examine the problem of adaptively scheduling perfect observations and preventive replacements for a multi-state, Markovian deterioration system with silent failures such that total expected discounted cost is minimized. We model this problem as a partially observed Markov decision process and show that the structural properties of the optimal policy hold for certain non-extreme sample paths.

show abstract

Section: Resultsmentioning

confidence: 99%

Structured maintenance policies on interior sample paths

Maillart

Zheltova

2007

Naval Research Logistics

Self Cite

View full text Add to dashboard Cite

show abstract

“…In the maintenance optimization literature, one of the key structural properties to establish is convexity of the value function (see, e.g., Dayanik et al 2008, Kim and Makis 2012, Maillart 2006, Makis 2008). In the current setting however, it is not clear that the robust value function W defined in (14) will preserve such a convex structure in the presence of positional and transitional model ambiguity.…”

Section: Structure Of the Decision-maker's Optimal Policymentioning

confidence: 99%

“…In the condition-based maintenance literature, there has been a long and rich history of establishing structural properties of optimal maintenance policies (see, e.g., Ross 1971, Makis and Jardine 1992, White 1978, Elwany et al 2011, Maillart 2006, Kurt and Kharoufeh 2010, Ulukus et al 2012. Although a lot of research has been conducted in this area, to our knowledge, this paper constitutes the first paper reporting structural results for maintenance optimization in the presence of model miss-specification.…”

Section: Introductionmentioning

confidence: 98%

“…A common modeling approach is to represent system degradation as an unobservable Markov process that can only be indirectly or partially observed using stochastically related signals collected through online condition monitoring (see, e.g., Dayanik and Gurler 2002, Dayanik et al 2008, Kim and Makis 2013, Maillart 2006. From the perspective of real-world implementation, this modeling approach implicitly assumes that the decision maker is able to accurately and completely specify the full joint distribution governing the real-world degradation/failure dynamics as well as the condition-monitoring observations.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Robust Control of Partially Observable Failing Systems

Kim

2016

Operations Research

View full text Add to dashboard Cite

This paper is concerned with optimal maintenance decision making in the presence of model misspecification. Specifically, we are interested in the situation where the decision maker fears that a nominal Bayesian model may be miss-specified or unrealistic, and would like to find policies that work well even when the underlying model is flawed. To this end, we formulate a robust dynamic optimization model for condition-based maintenance in which the decision maker explicitly accounts for distrust in the nominal Bayesian model by solving a worst-case problem against a second agent, "nature," who has the ability to alter the underlying model distributions in an adversarial manner. The primary focus of our analysis is on establishing structural properties and insights that hold in the face of model miss-specification. In particular, we prove (i) an explicit characterization of nature's optimal response through an analysis of the robust dynamic programming equation, (ii) convexity results for both the robust value function and the optimal robust stopping region, (iii) a general robustness result for the preventive maintenance paradigm, and (iv) the optimality of a robust control limit policy for the important subclass of Bayesian change point detection problems. We illustrate our theoretical result on a real-world application from the mining industry.

show abstract

“…A crucial part of the MDP model is the relative cost function that formulates the relative cost of a single step in the long-run decision process [14]. For the investigated maintenance optimisation problem, the relative cost function is given by:…”

Section: Markov Decision Process Modelmentioning

confidence: 99%