I agree with the previous post that a SPC chart could be used for tracking On-Time delivery. However, the acceptable tolerance should come from the "Customer" (i.e. Promise Date)otherwise you may be tracking a metric that means little to your customer.
I see no reason why you couldn't use spc to evaluate your shipping process. If you can measure it with variable data you can apply the SPC tools. You probably can define a target with acceptable limits that can serve as your tolerance - for example 4 weeks +/- 4 days after receipt of the order. You can then plot the actual results on an individuals chart to observe trends. If enough data is available (75-100 data points), you can evaluate for stability using Individual/mR or X-bar/R charting methods. Control limits can be applied and special causes identified, investigated and corected. If stable, then Cp/Cpk can be determined. However, this may take more time than you wish to accumulate the data. Another suggestion is to study smaller groups of data, say 25-30 orders, and calculate Pp/Ppk. The disadvantage of this is that stability is not considered. At any rate, I would not focus on the indices, but would attempt to understand the basline performance of the process so any changes made to improve delivery can be evaluated for effectiveness.
Community metadata Could not be loaded. No pre-configured community forum parent found for the incoming fid: . You need to add this forum into custom_code/community.sql
Send to friend
Comments
deroecrs 2/3/2004
I agree with the previous post that a SPC chart could be used for tracking On-Time delivery. However, the acceptable tolerance should come from the "Customer" (i.e. Promise Date)otherwise you may be tracking a metric that means little to your customer.
Rich DeRoeck
mkomarmy 1/29/2004
I see no reason why you couldn't use spc to evaluate your shipping process. If you can measure it with variable data you can apply the SPC tools. You probably can define a target with acceptable limits that can serve as your tolerance - for example 4 weeks +/- 4 days after receipt of the order. You can then plot the actual results on an individuals chart to observe trends. If enough data is available (75-100 data points), you can evaluate for stability using Individual/mR or X-bar/R charting methods. Control limits can be applied and special causes identified, investigated and corected. If stable, then Cp/Cpk can be determined. However, this may take more time than you wish to accumulate the data. Another suggestion is to study smaller groups of data, say 25-30 orders, and calculate Pp/Ppk. The disadvantage of this is that stability is not considered. At any rate, I would not focus on the indices, but would attempt to understand the basline performance of the process so any changes made to improve delivery can be evaluated for effectiveness.