# Analyzing Supplier On-Time Delivery

Six Sigma – iSixSigma Forums Operations Supply Chain Analyzing Supplier On-Time Delivery

Viewing 6 posts - 1 through 6 (of 6 total)
• Author
Posts
• #246861

James Barrett
Participant

I am trying to analyse our suppliers’ OTD, by no. days late. However of the 500-odd lines, the spread is 75% 0-6 days and the remainder a tail out to 500 days.

This makes for a crap histogram, control chart, time series plot, etc.

I’m looking for advice on a good way to analyse this, given every part that arrives late, by any amount, is significant.

0
#246877

Robert Butler
Participant

I would disagree with respect to your view of a histogram for delivery time.  I would insist it is just what you need to start thinking about ways to address your problem.

Given: Based on your post – just-in-time (JIT) for everything is a non-starter.

Under these circumstances the issue becomes one of minimization of the time spread for late delivery.

You know 75% of the parts have a time spread of 0-6 days.  What is the time spread for 80,90,95, and 99%?  Since all parts are created equal and all that matters is delivery time then the focus should be on those parts whose delivery time is in excess of say 95% of the other parts.  Once you know which parts are in the upper 5% (we’re assuming the spread in late delivery time for 95% of the parts is the initial target) the issue becomes one of looking for ways to insure a delivery time for the 5% is less than or equal to whatever the upper limit in late delivery time is for the 95% majority.

Having dealt with things like this in the past my guess is you will first have to really understand the issues surrounding the delivery times of the upper 5%.  I would also guess once you do this you will find all sorts of things you can change/modify to pull late delivery times for the upper 5% into the late delivery time range of the 95%.

Once you have the upper 5% in “control” you can choose a new cut point for maximum late delivery time and repeat.  At some point you will most likely find a range of late delivery time which, while not JIT, is such that any attempt at further reduction in late delivery time will not be cost effective.

2
#246885

James Barrett
Participant

Spectacular, thanks very much Robert.

0
#246900

Martinelli
Participant

As Robert said, to put your efforts in the 95% is the better way to improve your OTD…

We had a similar metrics at GE Plastics (we called as SPAM) this “SPAM” consist to sum (in module) the percentil 5% plus 95%, this means you remove 5% of your “best” numbers (that some time is not best, because if your supplier delivered before your needs you will have impact at you cash flow) and 5% of your “worst” numbers (some times special causes), we called as SPAM 90, and in the end our objetive is to reduce this SPAM. After we reaching the goals, we increse the SPAM to 95% (percentil 2,5% and 97,5%), 98% until 99%.

This SPAM were metrics of Supply Chain Area and the improve of this number were made by DMAIC projects leaded by supply chain BB.

0
#246902

hpsigmaguy
Participant

All great points. When your histogram, or other visual chart looks “crappy”, it’s not always that you picked the wrong analysis tool. It’s usually because your process is crappy somewhere.

I’m scoping several projects now that are similar dealing with response times. More times than not, our response times are anything from not good to down right terrible. Sometimes there are legitimate reasons for this so the data can not be discounted, but most of the time I suspect it’s just problems in the processes. The mess in the chart is my selling point to my stakeholders and sponsors as to why we need the project to begin with. I run a nice SPC chart and show them that the process is out of statistical control, therefore any improvement attempts would be futile. I usually get the go ahead to begin bringing my processes into control. Then I can pareto root causes and work on the big fish until I get to a point that any more projects on that processes would be cost prohibitive measured against any anticipated gains.

0
#246993

Rhb
Participant

You should not expect a histogram to be normally distributed.  Inherently, it will typically be skewed to the right and represent a Weibel distribution.  You could run a capability analysis  for non-normal data and set your upper boundary at the acceptable level for OTD.  This will give you a graphical picture of the process and an indication of your PPM failures.

0
Viewing 6 posts - 1 through 6 (of 6 total)

You must be logged in to reply to this topic.