Abstract

Machines are required to manufacture simple fans to complex ships, for a variety of functions. One of the
elementary challenges presently faced in today’s industries is the identification of machine faults before a critical/
catastrophic level is reached. The machines should work to their fullest capacity for the intended period of usage
in optimum condition complying with all the recommended parameters. This would mean reduction in unforeseen
downtime, cost, and productivity and elimination of catastrophic failure altogether. To achieve these objectives an
optimum maintenance plan has to be devised based on the type of industry machines used, funds available, quality
of finished products, etc. Bearings are most the essential components of any machine, which exhibit number of
parameters such as speed, temperature, vibration, noise, wear, and tear. Any machining problem identified well
in advance by the variation in these bearing parameters over a period. A minor investment toward predictive
and preventive maintenance monitoring system can avert premature failures, machine degradation, unforeseen
stoppages, and production loss. This would result in increased profits, safety, proper working of the machine for
the intended duration of its life. This paper proposes an algorithm to predict bearing degradation using LabVIEW.