Abstract
Predictive Maintenance (PdM) solutions assist decision-makers by predicting equipment health and scheduling maintenance actions, but their implementation in industry remains problematic. Specifically, prior research repeatedly indicates that decision-makers often refuse to adopt the data-driven, system-generated advice in their working procedures. In this paper, we address these acceptance issues by studying how PdM implementation changes the nature of decision-makers’ work and how these changes affect their acceptance of PdM systems. We build on the human-centric Smith-Carayon Work System model to synthesise literature from research areas where system acceptance has been explored in more detail. Consequently, we expand the maintenance literature by investigating the human-, task-, and organisational characteristics of PdM implementation. Following the literature review, we distil ten propositions regarding decision-making behaviour in PdM settings. Next, we verify each proposition’s relevance through in-depth interviews with experts from both academia and industry. Based on the propositions and interviews, we identify four factors that facilitate PdM adoption: trust between decision-maker and model (maker), control in the decision-making process, availability of sufficient cognitive resources, and proper organisational allocation of decision-making. Our results contribute to a fundamental understanding of acceptance behaviour in a PdM context and provide recommendations to increase the effectiveness of PdM implementations.
Original language | English |
---|---|
Pages (from-to) | 7846-7865 |
Number of pages | 20 |
Journal | International Journal of Production Research |
Volume | 61 |
Issue number | 22 |
DOIs | |
Publication status | Published - 2023 |
Keywords
- Behavioural operations management
- decision support systems
- human factors
- human–computer interaction
- Industry 5.0
- predictive maintenance
Fingerprint
Dive into the research topics of 'Predictive maintenance for industry 5.0: behavioural inquiries from a work system perspective'. Together they form a unique fingerprint.Press/Media
-
Onderzoek TU/e: alleen mensen kunnen zorgen dat algoritmes vertrouwd worden
16/02/23
1 Media contribution
Press/Media: Research
Datasets
-
Predictive maintenance for industry 5.0: behavioural inquiries from a work system perspective
van Oudenhoven, B. (Creator), van de Calseyde, P. P. F. M. (Creator), Basten, R. J. I. (Creator) & Demerouti, E. (Creator), Taylor and Francis Ltd., 15 Dec 2022
DOI: 10.6084/m9.figshare.21731357
Dataset