Metrics and Games
Abstract
Carl and Fred discussing a listener question on maintenance strategies, and a company mandate metric comparing preventive and corrective maintenance tasks.
Key Points
Join Carl and Fred as they discuss how metrics can become a numbers game if they are used incorrectly.
Topics include:
- Sometimes a metric can alter behavior in the wrong way if people learn to “game” the system.
- How metrics can become a numbers game, especially if too much force is used to meet a threshold.
- Using RPN thresholds, as an example
- Ratio of corrective to preventive tasks is vulnerable to numbers game.
- Ask: what you are really looking for? How do we achieve that?
- The *quality* of corrective and preventive tasks is essential.
- RCM can be used to get the right tasks; needs the right inputs and criticalities.
- How do you encourage more preventive maintenance in a meaningful way.
- What are quality objectives for an RCM project? How can these be used?
- What are quality objectives in an FMEA project and how are they used?
- Trusting the team vs monitoring data
- What is role of Root Cause Analysis in this discussion?
- What can you do to improve maintenance procedures when in middle of production?
- Effectiveness of corrective actions
- How to implement future preventative actions
- Use of 80/20 rule in prioritizing tasks
Enjoy an episode of Speaking of Reliability. Where you can join friends as they discuss reliability topics. Join us as we discuss topics ranging from design for reliability techniques to field data analysis approaches.
/
RSS Feed
Related Topics
Corrective Action Question(Opens article in a new browser tab)
Thank you. Some interesting points raised. Keep up the good work.
Thanks, Carl. We appreciate your feedback.
The social scientists Donald Campbell and Charles Goodhart both arrived at the idea “When a measure becomes a target, it ceases to be a good measure.” The manager dictating the 70/30 split created a game and everyone got onboard to play.
I think most managers fall victim to this thinking–I see what the measure says and I would like it to be something specific. This leads to the corruption we see with generating data (MTBF, OEE, GRR, US News & World Report school rankings, any other quality and performance metric you can think of) to get the specified outcome.
A particularly egregious example was the deployment of Six Sigma in one division in the corporation where I worked. The leadership said anyone who didn’t get certified would be fired. Since there weren’t any mass layoffs, I randomly picked a few projects in the project tracker database to see how good the reports were. As you could imagine, the project reports were just garbage. Leadership got its certifications, though.
Thanks for your comments, Keith. I love the quote: “When a measure becomes a target, it ceases to be a good measure.” This applies to so many reliability areas, such as RPN, Control Charts and many more. The Six Sigma example you shared highlights this problem very well.