There has been a lot of discussion as to why we are asking technicians to write corrective work orders for problems found on a PM instead of fixing the issue on the spot. If we consider the example of the four machines below, it becomes clear why this is the case.
A technician is assigned a 2 hour PM for each of the machines below. As he checks #1, there are no issues found and the PM is completed with no corrective work order generated. As the inspection of #2 begins, he notices an issue. This issue is not very critical and could wait until next weekend. He decides to fix it anyway. The repair takes another 2 hours. He moves on to machine #3. There are no issues that demand immediate attention. Since the PM of #2 and the corrective work order to fix the issue have taken a total of 4 hours, it is now time to go home, and the PM for machine #4 does not get done.
The problem is, machine #4 has a failure in it that will shut the machine down in a day if not detected. The next shift coming in has it’s own list of things to do and will not get to the PM on machine #4 that the previous shift did not complete.
What we are trying to accomplish is to get as many failures detected as early on in the failure curve as possible. It is understood that some repairs need to happen as soon as the fault is detected. Most of the time this is not the case. With the chance to plan for a repair, all of the parts can be ordered and the job will take less time. We need to adapt to more of an inspection mindset when doing a PM and provide a clearly written corrective work order when a failure is detected. This is not the way things have been done in the past. We cannot continue to do things the same way and expect better results. We must change the way we operate in order to improve.
Great example! We hope this helps you with a topic to share or can be a model for a similar communication to your group as you chase after reliability.
We have two PM's where I work; I know because I write them!
ReplyDelete1. Inspect and Check
2. Replace
#1 is what it sounds like - inspect for key items that are either OEM-driven, or which derive from historical experience.
We include in this process safety and certain quality checks, as well.
#2 is a refurbishment-based process. We have not gone completely "condition based," so some items are simply replaced and/or rebuilt at periodic intervals. This carries a bit more cost, but we do it very strategically. In our business (pharma), it is better to err on the side of safety.
The last sort of work order is the non-routine. These can be emergent, or driven from the PM process.
Each PM has incorporated into it instructions to repair/replace minor items, but to capture crucial, or more involved work that cannot be done during the PM on a follow up, non-routine work order.
I can understand this procedure being implemented in a production setting where equipment is being serviced during down times under a controlled schedule. The procedure appears to have some merit. I have a problem using your model in a facilities operations setting like a hospital or Pharma environment. For example, I work for an outsource company doing operations for a facility with over 2 million square feet of space. Assigning PM's and not staffing enough head counts for daily occurrences has created a conflict between the way we used to do things in the past and the new way companies are trying to run their maintenance operations. Any comment in this regard?.
ReplyDeleteThat's the opinion for using this application on a manufactures operation on a production line in a Nissan plant. I am curious to know what the opinion is when this model is used in other parts of the industry. I have less than favorable opinion of this method and found the process
ReplyDeletenot so effective when applied to an outsource operations group at a R&D facility where day to day activities are less structured than at an assembly or manufacturing operation.
Finally, I would encourage you to take a look at the precision maintenance skill set of the technicians who conduct the repairs. If you’re not leveraging tools such as alignment and belt tension, to name a few, you might be inducing problems that will lead to premature failure of the asset and feeding the find/fix loop.
ReplyDeleteNot every follow up to a PM will always be planned and scheduled but we should be striving to find defects as early as possible to enable the proactive process. For the work that can’t or shouldn’t be planned and scheduled based on business needs and constraints we should have a process to manage it and capture reliability data for trending and improvement.