Regression for MTBF (Medical Device: Repairs vs Install Base) with Reliability Data

Romvill2002

Involved In Discussions
Sorry to throw you guys off the topic but I need help on what should be the best regression model for MTBF(medical device, repairs vs install base)???

Install Repairs
675 15
872 16
1176 18
1353 29
1575 27
1879 46
2035 48
2302 41
2837 63
2961 53
3132 65
3266 100
3425 88
3597 98
3941 125
 

Miner

Forum Moderator
Leader
Admin
Re: Regression with reliability data

This is not MTBF prediction, but the relationship between the installed base and repairs,

Can you provide more information? Is the the installed base increasing over time? More data and context would yield a better response.
 

Attachments

  • Regression for Repairs vs Install - Prediction Report.jpg
    Regression for Repairs vs Install - Prediction Report.jpg
    95.7 KB · Views: 251

Romvill2002

Involved In Discussions
Re: Regression with reliability data

Correct, not trying to predict MTBF but just wanna know if linear or quadratic regression is best to graphically represent the data
 
S

supreecha

Re: Regression with reliability data

Fitted Line Plot :
 

Attachments

  • Regression Model.jpg
    Regression Model.jpg
    59 KB · Views: 216

Romvill2002

Involved In Discussions
Re: Regression for MTBF (Medical Device: Repairs vs Install Base) with Reliability Da

Thanks, looks like quadratic is the best type.
 

Bev D

Heretical Statistician
Leader
Super Moderator
Re: Regression for MTBF (Medical Device: Repairs vs Install Base) with Reliability Da

what are you trying do? why regression? if you look at the control chart it is clear that this process has two stable periods. something clearly changed at sample 12.

My concern is that you may be making this far too complicated?
 

Attachments

  • regression or control chart.pptx
    75.1 KB · Views: 135

Romvill2002

Involved In Discussions
Re: Regression for MTBF (Medical Device: Repairs vs Install Base) with Reliability Da

Im trying to find how strong the relationship of service events with the accumulative installed base. Then will try to figure out which control chart should I use. You have chosen a P-chrt and out of control in the sample 12 due to high returned devices for repairs.Should i just use IMR chart or p-chart?
 

Bev D

Heretical Statistician
Leader
Super Moderator
Re: Regression for MTBF (Medical Device: Repairs vs Install Base) with Reliability Da

not sure what the relationship between the install base and the count of repairs will tell you without first looking at the time series.

statistical inference and model building do rely on the homogeneity of the underlying process and with the shift in failure rates at point 12 you do not have a homogenous process so any interpretation of the regression plot alone will be misleading as the shift is confounded with (or coincident with) a particular point in time when the install was increasing. for example, if the defect rate is constant, then an increase in the install base will result in a an increase in the number of repairs and we would expect that there would be a linear relationship between install base and repairs. Now, let's say that at point 12 there was a new defect introduced with a new batch of one of the system components. IF this were the case imagine what the defect rate would have been in points 1-11 if those instruments had the 'new defect rate': it would be higher. so the curve in the regression plot is NOT due to the install base as much as it is due to the 'new defect' introduced in point 12.

There are of course other complications that could shape the regression curve. another example is if the repaired units are returned to the install base their defect rate will change and now you have a mix of instruments that will have different failure rates. this is why we often do a 'vintage' time series to pull this apart...

fancy statistics can be very seductive, but I've found that the simple stuff - plot your data in it's rawest forms in time series and ask yourself questions about that first - is a far more productive approach. there are so many nuances about processes that can get hidden by the 'blind' application of statistical tools. If you first understand the nature of your data and the process it represents you will not be blind when you select a more sophisticated statistical tool. I tell my students to imagine what would happen if they were to pick up a chainsaw to cut down a tree but they were blindfolded. :mg:

As for which chart to use, you need to try both and determine which one fits your need best. *I* have found that the p chart works reasonably well with field repair/return/warranty data especially when the install base is growing substantially as it modifies the control limits based on the install base size and this type of data is reasonable modeled by the binomial. the I, MR chart doesn't do that. I have attached a new file that shows both charts for your data; the upward shift at point 12 is still clear in I, MR chart...
 

Attachments

  • regression or control chart.pptx
    99 KB · Views: 137

Romvill2002

Involved In Discussions
Re: Regression for MTBF (Medical Device: Repairs vs Install Base) with Reliability Da

Hi Bev,

Thank you so much. I used the IMR chart, since the device is newly introduced to the market and due to ramp up on sales installed base rate is aggressively inclining. Maybe when both service and install matured then I may use the pchart. What do you think?
 

Bev D

Heretical Statistician
Leader
Super Moderator
Re: Regression for MTBF (Medical Device: Repairs vs Install Base) with Reliability Da

given that I, MR charts doesn't adjust for sample size (install base), *I* would do the opposite. This works much better in my experience. p chart first then I, MR chart when the install base stabilizes, it ever. UNLESS you can determine that the underlying process is a large departure form the binomial distribution and I've not seen that in 30 years of doing this...
 
Top Bottom