Optimizing Predictive Maintenance Schedules
Below is a step by step example of how to optimize predictive maintenance
schedules. The example utilizes the Reliability & Maintenance
Analyst software. Click here to download a free demo version of
Click here for the technical details of optimizing schedules for preventive maintenance, predictive maintenance, and for optimizing inspection schedules in Adobe Acrobat format. For the same article in Microsoft Word for Windows 95 format click here.
A vibration analysis is conducted on a fan once per month. Vibration readings beyond a specific level indicate an impending failure. The goal is to do vibration analysis often enough to catch the machinery in a state of high vibration, but before failure. This state is referred to as the lapse zone. The length of the lapse zone must be estimated. It is not feasible to run to failure for the purpose of collecting failure data. The length of the lapse zone is determined by estimating a minimum length, a maximum length, and a most likely length. For this example, production operators determined that the minimum duration of high vibration reading before a failure occurs is 3 million feet of production, the maximum duration is 10 million feet of production, and the most likely duration is 5 million feet of production.
The cost of a vibration analysis is $22, and the cost of a machinery failure is $985. The time to fail data in this case should be the time until increased vibration is first noticed. The file "VBRATION.DAT" (This file is included with the demo version.) contains data for the time until vibration readings began increasing. How often should vibration readings be taken?
The software will display the results in a frame. The first vibration analysis should be conducted after 88.23 million feet of production have been run. The second vibration analysis should be conducted after 92.11 million feet of production have been run, only 3.88 million feet later.
Note that the time between each successive inspection is decreasing. This is because the failure rate is increasing. A shape parameter greater than 1.0 (the shape parameter was 4.503 for this example) indicates "wear-out", which is characterized by an increasing failure rate. The longer the fan runs, the more likely it is to fail. If the shape parameter is equal to 1.0, the time between each successive inspection will be equal. If the shape parameter is less than 1.0, the time between each successive inspection will increase; the longer the item runs the less likely it is to fail.
Engineered Software, Inc.