Condition‐based maintenance is the diagnosis of component failure or a prognosis of a component's time to failure. The aim of this paper is twofold: a summary of the main assumptions regarding condition‐based maintenance found in the literature into eight postulates, and a comparison of the postulates against industrial practice. The postulates were formulated regarding the technical system, the managerial system and workforce knowledge.
The postulates were examined in a multiple case study of five large firms in the process industry.
The results indicate that some postulates were supported with empirical findings. Limited or no support was found for postulates concerning the application of prognostic activities, use of dedicated software, use of procedures, use of training, and the active management of domain‐related knowledge availability.
Practitioners can use the eight postulates as key elements in the management of condition‐based maintenance technology, and for the comparison of their current condition‐based maintenance practices with what literature generally proposes.
Other researchers have reported on condition‐based maintenance, but most publications focus on applied mathematics and new monitoring and simulation models. Only limited attention was paid to industrial practice so far. The study is one of the first in‐depth empirical studies into actual condition‐based maintenance practice.
