Abstract:
Lubrication Condition monitoring (LCM) is not only utilized as an early warning system in
machinery but also, for fault diagnosis and prognosis under condition-based maintenance
(CBM). LCM is considered as an important condition monitoring technique, due to the
ample information derived from lubricant testing, which demonstrates an introspective
reflection on the condition and state of the machinery and the lubricant. Central to the
entire LCM program is the application concept, where information from lubricant analysis
is evaluated (for knowledge extraction) and analyzed with a view of generating an output
which is interpretable and applicable for maintenance decision support (knowledge appli-
cation). For robust LCM, varying techniques and approaches are used for extracting, pro-
cessing and analyzing information for decision support. For this reason, a comprehensive
overview of applicative approaches for LCM is necessary, which would aid practitioners
to address gaps as far as LCM is concerned in the context of maintenance decision support.
However, such an overview, is to the best of our knowledge, lacking in the literature, hence
the objective of this review article. This paper systematically reviews recent research
trends and development of LCM based approaches applied for maintenance decision sup-
port, and specifically, applications in equipment diagnosis and prognosis. To contextualize
this concern, an initial review of base oils, additives, sampling and testing as applied for
LCM and maintenance decision support is discussed. Moreover, LCM tests and parameters
are reviewed and classified under varying categories which include, physiochemical, ele-
mental, contamination and additive analysis. Approaches applicable for analyzing data
derived from LCM, here, lubricant analysis for maintenance decision support are also clas-
sified into four categories: statistical, model-based, artificial intelligence and hybrid
approaches. Possible improvement to enhance the reliability of the judgement derived
from the approaches towards maintenance decision support are further discussed. This
paper concludes with a brief discussion of plausible future trends of LCM in the context
of maintenance decision making. This present study, not only highlights gaps in existing
