Minitab tool to evaluate PM (Preventive Maintenance) process

#1
Hello,

I just want to seek an advise from you guys about a study that I am working on.
I need to evaluate the effectiveness of preventive maintenance process using statistical tools in minitab.
Given data are the following:
1. Number of defectives for Assembly Line 1 per day
2. Number of defectives for Assembly Line 2 per day
3. Cumulative data counter at the time of a PM cycle
I have attached a sample data.

I am confused what statistical tool or graph to use to stratify the data (and perhaps to do a test?) and show how effective the PM cycle for Assembly Line 1 and 2.
Do you have any ideas using the given the data.

I tried P-chart but I am not sure if it is appropriate.

Thanks,
miongski
 

Attachments

Last edited:

Miner

Forum Moderator
Staff member
Admin
#3
I think a p-chart is an appropriate tool to use. You can split the worksheet into 2, one for each production line. Calculate your limits, excluding out-of-control data. I noticed that these out-of-control points seemed to be followed by maintenance, whereupon, the chart was back in control. Are you certain that these are PMs, or are they corrective maintenance?
 
#4
Hi Miner,

I missed to include this information, PM should be done every 4000 parts processed.
I used the P-chart as shown below, I used the "Parts Processed" data as Subgroup Size instead of "Cumulative Parts Processed Counter" (is this correct?). The out of control points seems to be from PMs that was made beyond the 4000 parts.
So you suggest to remove these 4 out of control points, then calculate control limits, below is the p-chart showing the out-of-control points.


1555955227131.png


Below is the p-chart after removing the 4 out-of-control points, now its showing new 2 out-of-control points. Will I remove this 2 points again then create another p-chart using the UCL and LCL below?

1555955764275.png
 
#5
I think a p-chart is an appropriate tool to use. You can split the worksheet into 2, one for each production line. Calculate your limits, excluding out-of-control data. I noticed that these out-of-control points seemed to be followed by maintenance, whereupon, the chart was back in control. Are you certain that these are PMs, or are they corrective maintenance?
Hi Miner,

I missed to include this information, PM should be done every 4000 parts processed.
I used the P-chart as shown below, I used the "Parts Processed" data as Subgroup Size instead of "Cumulative Parts Processed Counter" (is this correct?). The out of control points seems to be from PMs that was made beyond the 4000 parts.
So you suggest to remove these 4 out of control points, then calculate control limits, below is the p-chart showing the out-of-control points.



1555955227131-png.png



Below is the p-chart after removing the 4 out-of-control points, now its showing new 2 out-of-control points. Will I remove this 2 points again then create another p-chart using the UCL and LCL below?


1555955764275-png.png
 

Miner

Forum Moderator
Staff member
Admin
#6
You are on the right track. It is a bit of a judgement call on when to stop. I suspect that when you remove the two red points plus the 3 close to the limits, you will have a good estimate of p-bar. You can then set p-bar to that level under Options > Parameters > proportion and include all of the data points.
 
#7
You are on the right track. It is a bit of a judgement call on when to stop. I suspect that when you remove the two red points plus the 3 close to the limits, you will have a good estimate of p-bar. You can then set p-bar to that level under Options > Parameters > proportion and include all of the data points.
thanks for your help. I will try it.
 

Similar threads



Top Bottom