Embedded Files
What is Condition based Preventive Maintenance (CbPM)?
Condition-based maintenance scheduling, often referred to as CBM scheduling, is a proactive approach to maintenance management that relies on real-time data and advanced monitoring technologies to assess the actual condition of equipment. Various sensors are strategically placed on equipment and machinery, collecting data on various performance metrics such as temperature, vibration, pressure, and fluid quality. By continuously analyzing the health and performance of assets, CbPM allows businesses to schedule maintenance activities precisely when they are needed, minimizing downtime, reducing maintenance costs, and maximizing operational efficiency. This data-driven approach empowers organizations to make informed decisions, prioritize maintenance tasks, and extend the lifespan of their assets, ultimately enhancing reliability and productivity.
Model
The life of the equipment is modeled as a Multi State Markov System with 2d states (Figure 1).
State 1: Initial state (perfect condition)
State (2j-1): Degraded state; j = 2, …., d
State (2j): Poisson failure state; j = 1, …., d - 1
State 2d: Complete failure state
The state probabilities Pi(t), i=1,…., 2d, are obtained by solving the system of Chapman–Kolmogorov equations corresponding to the Markov model.
Repair rates are taken to be varying exponentially based on literature review.
Figure 1: System State Transition Diagram
Total cost is modeled as sum of three types of cost:
1] Perfect repair cost: It is equal to number of times the system undergoes perfect repair multiplied with the cost of each perfect repair
2] Imperfect repair cost: After a poisson failure, the equipment is repaired minimally to the last working state
3] Loss of production cost: It is the loss of production because of maintenance downtime. It is not a physical monetary cost, that is, it is the loss of revenue when the equipment is not working according to its normal schedule.
Data Processing
The sensors' data was divided into d working states by K means clustering.
K Means Clustering: It divides the population or set of data points into a number of groups so that the data points within each group are more comparable to one another and different from the data points within the other groups. ‘K’ in the name of the algorithm represents the number of groups/clusters we want to classify our items into. Euclidean distance is used as a measurement of similarity.
Results
The state of minimal total repair cost was identified as the 7th stage of degradation. This outcome stems from three cost components, one of which displays a different trend compared to the other two. The cost of production loss diminishes over time: initially, a perfect repair during the machine's high efficiency phase leads to a significant production loss. However, as the machine deteriorates, the cost of production loss decreases. Perfect repair costs increase with the machine's degradation since, in the initial stages, fewer repairs and lower repair costs are incurred when the equipment is nearly perfect.
Cost components vs Degradation state
Page updated
Google Sites
Report abuse