The business case for Industrial Analytics for Predictive Maintenance is intuitive and relatively easy to justify. There are numerous studies and data to demonstrate the potential revenue gained by increased uptime across a wide range of industries. The most effective way to secure executive buy-in for an investment decision is to forecast positive top-line growth.
However, a thorough ROI based on the economic benefits of industrial analytics require a more granular analysis of operational metrics. This article reviews the impact of industrial analytics for predictive maintenance on wrench-time.
What is wrench time and why is it important?
Wrench time refers to the amount of time that maintenance workers are actively engaged on the job with “tool-in-hand.” The calculation of wrench time excludes all other activities of maintenance employees including:
- Traveling/transportation to the repair site
- Delays for parts, rental equipment or relevant documentation
- Repair instructions
- Waiting for machinery to be shut down
- Waiting for additional support to arrive at the job site
- Preparations for repair and clean-up
With careful planning and insights into the root cause of machine failure, a significant amount of non-wrench time activities can be reduced.
Step-by-step guide for calculating wrench time improvements
It is estimated that as little as 30% of a repair worker’s workday is dedicated to wrench time. Based on an eight-hour workday, that translates into 2 ½ hours per day. The economic cost of non-wrench time to a production facility is significant. Here are some basic calculations:
Let’s assume a 40 hour / 52 week of work for a repairperson with an average hourly cost of €30. The daily cost per repairperson is €240 based on an 8-hour workday. If 2 ½ hours are allocated to wrench activities, then 5 ½ hours are allocated to non-wrench activities. Based on the hourly cost of €30, that translates into €75 per day in wrench activities and €165 in non-wrench activities. If the full 8-hours are allocated to wrench activities, the cost allocation to wrench activities is €240 per day.
If wrench time can be increased from 30% to 50%, then this frees up 31,200 labor hours at the cost of €936,000 for a facility with a 50-person maintenance crew and 124,800 hours or €3,744,000 for a 200-person maintenance force.
How does Industrial Analytics for Predictive Maintenance Improve Wrench Time?
There are two aspects to Industrial Analytics that should be considered.
1) Root Cause Analysis
With solutions such as SKF Enlight AI’s Machine Learning for Asset Maintenance, the learning algorithm detects abnormal patterns of sensor behavior or correlations of these patterns. Identifying the sequence of abnormal sensor activity provides insight into the Root Cause Analysis.
How is this relevant to wrench time? Part of the challenge for maintenance workers is identifying why a machine is failing or has broken down. This requires significant guesswork, especially for relatively untrained technicians. If a breakdown can be traced to the behavior of a specific sensor in a piece of equipment, this helps identify the specific repair activity that is required for remediation.
2) Time-to-Failure Machines
Using Big Data for Predictive Maintenance provides earlier indications of degradation or asset failure. For instance, at a wind turbine facility, alerts were triggered by SKF Enlight AI over 50 hours before expected downtime. Non-wrench time activities can be reduced or eliminated when there is extra time for upfront planning. Repair workers’ time can be allocated to other tasks instead of requiring them to wait unproductively for parts or extra support to arrive. Finally, the need for overtime can be reduced with improved scheduling.
Summary and Conclusion
A comprehensive business case should be based on both top-line and bottom-line considerations. Productivity metrics are often overlooked in calculating the ROI for Industrial Analytics for Predictive Maintenance. Improved operational efficiency from increasing wrench time can result in significant costs savings.