Prognostic modelling for industrial asset health management

Abstract

Failure prognostics and health management are central to the Remaining Useful Life (RUL) estimation of critical engineering assets, particularly to improve safety, reduce downtimes and maintenance expenditures. Over recent years, several prognostic approaches have been developed to predict remaining asset lifetime, optimise maintenance schedules, and enhance equipment availability and reliability. While academic research in this area has grown rapidly, implementations of these methods by industry’s asset managers and reliability experts have only had limited success. Yet asset lifetime and reliability analysis are only restricted to the conventional reliability-centred maintenance and total productive maintenance approaches in industries. The purpose of this paper is to emphasise a need for a paradigm shift in industrial asset health management from the conventional to modern approaches that would benefit industries. At first, this paper classifies existing prognostic techniques into the traditional reliability, model-based, and data-driven approaches. Each prognostic approach is then analytically discussed with emphasis on models and algorithms. Consequently, this paper explores the strengths and weaknesses of main models in these groups to assist industry practitioners to select an appropriate prognostic model for RUL prediction within their specific business environment. Finally, the paper concludes with a brief discussion on possible future trends and further research directions in this field.

Keywords:

• Asset health management
• Condition-based maintenance
• Fault diagnostics
• Failure prognostics
• Reliability modelling

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been republished with minor changes. These changes do not impact the academic content of the article.