Maintenance Management Optimization to Improve System Availability Based on Stochastic Block Diagram

SUMMARY & CONCLUSIONSOne of the most adopted strategies by the industrial sector for asset management is the monitoring of Key Performance Indicators (KPIs), which contribute to determine whether an organization is achieving its predetermined objectives. Amongst the most widespread, availability is an important KPI in the management of an asset, since it expresses its percentage of actual uptime relative to the total service hours. There are several factors that can affect the availability of an asset, such as its design characteristics, the mode of operation, the medium in which it is used, and the quality and agility of the maintenance actions practiced in it.Several techniques have been developed over the years to estimate asset availability. Among them, the Reliability Block Diagram (RBD) has become one of the most used, as it manages to determine, through a simple graphical representation, the reliability, availability and maintainability of repairable systems. However, this tool has a number of limitations regarding the dynamic behavior of the system under study, as it does not allow factors such as the size and workload of the maintenance team to be taken into account, nor the presence (or absence) of spare parts in stock.With this in mind, this work aims to use Stochastic Block Diagram (SBD), which combines the graphic representation of RBD with Monte-Carlo simulation, to estimate and evaluate the availability of a dynamic system. In the proposed method, factors such as the supply and stock of spare parts, as well as the size and workload of the maintenance team, are taken into account for a more precise estimate of the availability of the system under study.As a case study, the method is implemented in a hydroelectric plant. The results obtained from the plant model allow the optimization of the size of the maintenance team and the number of spare parts in order to improve system availability. The proposed method proved to be capable of assisting maintenance management regarding decision making to increase maintenance performance.