A static approach to investigate the impact of predictive maintenance in the reliability level and the failure cost of industrial installations

Digital IoT(Internet of Things)solutions for equipment condition monitoring andnew advanced algorithms to process big data, enablethe application of predictive maintenance.Consequently, actual implementations of such a system in industrial installations triggers theverification of its potentialbenefits. Thus, this project attempts to quantify the impact of a predictive maintenance system in the failure rate and the maintenance costof industrial installations.The lack oftime depended data lead to a static approach that utilizes average failure rate and mean time to repair values coming from IEEE standards and other sources. Next, a methodology that links the equipment causes of failure with a predictive maintenance system functions, is proposed. Consequently, new reduced failure rates for theassets under monitoring are defined.To perform the reliability calculations the spreadsheet methodology is presented and utilized. Additionally, the revenue requirement methodology is described and is used for the cost benefit analysis.Finally, the approach is applied in two theoretical and two actual industrial installations. Sensitivity analyses regarding different parameters of a predictive maintenance system are conducted in the first two cases,to evaluate the impact on different reliability indices. Moreover, cost benefit analysis is performed in the actual industrial networks and according to the resultspredictive maintenance should be preferred. Lastly, regarding the failure rate, a small or high reduction is observed depending on the type of failures, the utility sources,the system configuration,the number of monitored equipment and other paramet