A hierarchical decision-making framework for the assessment of the prediction capability of prognostic methods

Zhiguo Zeng, Francesco Di Maio, Enrico Zio and Rui Kang

Journal of Risk and Reliability, 2017, vol. 231, issue 1, 36-52

Abstract: In prognostics and health management, the prediction capability of a prognostic method refers to its ability to provide trustable predictions of the remaining useful life, with the quality characteristics required by the related maintenance decision making. The prediction capability heavily influences the decision makers’ attitude toward taking the risk of using the predicted remaining useful life to inform the maintenance decisions. In this article, a four-layer, top-down, hierarchical decision-making framework is proposed to assess the prediction capability of prognostic methods. In the framework, prediction capability is broken down into two criteria (Layer 2), six sub-criteria (Layer 3) and 19 basic sub-criteria (Layer 4). Based on the hierarchical framework, a bottom-up, quantitative approach is developed for the assessment of the prediction capability, using the information and data collected at the Layer-4 basic sub-criteria level. Analytical hierarchical process is applied for the evaluation and aggregation of the sub-criteria and support vector machine is applied to develop a classification-based approach for prediction capability assessment. The framework and quantitative approach are applied on a simulated case study to assess the prediction capabilities of three prognostic methods of the literature: fuzzy similarity, feed-forward neural network and hidden semi-Markov model. The results show the feasibility of the practical application of the framework and its quantitative assessment approach, and that the assessed prediction capability can be used to support the selection of the suitable prognostic method for a given application.

Keywords: Prognostic and health management; remaining useful life; prediction capability; analytical hierarchical process; fuzzy similarity; feed-forward neural network; hidden semi-Markov model (search for similar items in EconPapers)
Date: 2017
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://journals.sagepub.com/doi/10.1177/1748006X16683321 (text/html)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:sae:risrel:v:231:y:2017:i:1:p:36-52

DOI: 10.1177/1748006X16683321

Access Statistics for this article

More articles in Journal of Risk and Reliability
Bibliographic data for series maintained by SAGE Publications ().