1. Capture Reasons
It is essential to capture a reason and a duration for each downtime event so your team can effectively interpret the data and prioritize actions. In order to create actionable data:
- Start simple, with no more than 25 reasons, one of which should be All Other Losses.
- Make sure every reason is clear on its own, as well as when compared with other reasons.
- Prefer reasons that describe symptoms as opposed to root causes so that analysis and conclusions are left in the domain of CI and engineering.
- Remove reasons that aren't regularly used, and add reasons as needed to ensure that All Other Losses is not in the top ten losses.
- Add additional detail in the form of comments for especially long downtime events.
2. Perform Hourly Reviews
Every hour, hold a two-minute standup meeting at the line to review the longest downtime event (or other issue) from the last hour. Then, agree on one improvement action for the next hour. This is a simplified version of Short Interval Control (SIC). Read more about SIC.
3. Separate Quick Fixes and 100-Year Fixes
Make a conscious decision to move forward with each downtime loss in one of two ways:
- Quick Fix: Move forward with an immediate fix that can be implemented without outside resources.
- 100-Year Fix: Escalate a problem that needs outside resources to management for a permanent fix.
4. Treat Downtime as a KPI
Metrics that are emphasized and shared are very powerful drivers of behavior. Treat downtime as a KPI (Key Performance Indicator), and continually reinforce its importance. People love an opportunity to win, which is why we recommend a TAED scoreboard for the plant floor (Target, Actual, Efficiency, Downtime). This enables operators to track their progress in real time - and “win the shift.”
5. Change One Thing at a Time
When fixing any loss, make one change at a time. Often, multiple changes are made at the same time without fully understanding their impact. This makes it much harder to evaluate the effectiveness of each individual action.
6. Focus On the Constraint
Every manufacturing process has a constraint (i.e., bottleneck), which should be the focus of your improvement efforts. It is the point of leverage for the entire process. Measure downtime at the constraint (by automatically monitoring cycles), and Improve the Constraint to ensure that your resources are focused where they will have the greatest impact.
7. Make Down Events Visual
Provide clear visuals to indicate when a line is down, and escalate the message if the line remains down for an extended period of time. Train team members to react quickly, and provide multiple levels of escalation (e.g., operator, supervisor, manager). The goal is to prevent small issues from becoming large incidents, and if they do become larger incidents, to react quickly.
8. Perform a 3S Blitz
A clean and organized work environment creates better conditions for well-running equipment, which is why 5S is the foundation of TPM. A 3S blitz applies the same principles as 5S in the form of a one-off event: Sort (eliminate what is not needed), Straighten (organize what remains), and Shine (clean and inspect equipment).
9. Perform a Maintenance Blitz
Most equipment includes at least some parts that wear (e.g., seals, gaskets, bearings, belts, and rollers), which can cause breakdowns as these parts reach the end of their operational life. Check that all wear parts are in good condition, and replace any parts that are suboptimal.
10. Mark Optimum Settings
When equipment settings are continually tweaked, they are less likely to be optimal and more likely to create conditions that increase downtime. It is surprising how often this happens. Determine optimal settings as part of standard work and mark them on the equipment, making it is easy for your plant floor team to verify that the settings are as intended.
Some of the descriptions and graphic materials are from https://www.vorne.com/.
