Leaderboard

Any Manufacturing company should produce product your customers want to pay for. To find out that your process are capable you should use Cp, Cpk,Pp,and Ppk. Capability indices like Cp and Cpk are related to Customer specification whereas Performance Indices are related to Process Specifications. Capability indices are related to engineering specification determined by customer to observed behaviour of process. Capability of process is defined as the ratio of the distance from process center to the nearest specification limit divided by a measure of the process variability. Cp =a Simple indicator or Process capability. Cpk= Adjustment of Cp for the effect of non centered distribution. Cpk measure how close you are to target or specification limit and your consistent around your average performance. a person may be performing with minimum variation but away from specification limit may have high Cp. Also even if person is performing near specification limit with high variation may have high Cp Cpk will be higher only when you are meeting target or specification limit with minimum variation. As per standard you should have Cpk of 1.33 ( 4 Sigma) to satisfy most of the customers. Pp= a simple indicator of Process performance. Ppk= Adjustment of Pp for the effect of non centered distribution. Ppk Measures sample generated fro the process is capable to meet customer specification or requirement. Ppk is used when process control can not be evaluated for example pre-production run. Difference between Cpk and Ppk. Cpk tells you the process is CAPABLE of doing in future, assuming it remains in state of statistical control. So it is for short term. Ppk tells you the how the process has performed in the past. It is more realistic. So it is for Long term. While using DMAIC, it is more realistic to use use Pk and PPk than Cp and Cpk as the process variation can not be tampered with by inappropriate subgrouping.

Cp, Cpk, Pp & Ppk are used to see if the process is capable. This concept was widely used in Automobile industry as they had series production (600-700 child parts in day) with functional critical & significant characteristics they realized blindly cannot rely on just 100% Inspection of parts. 100% inspection of all the parts and all the dimensions is not practical for them as numbers were big. So they needed a way to get away with sampling inspection. Cp = A Capability index, Compares the process variation to the maximum allowable variation as indicated by the tolerance. Process variation is derived from within the subgroup range (Sometimes called as short term process capability). Cp Does not consider process location. Calculated as follows Cp = USL-LSL / 6 Sigma Where Sigma = R/D2 R = Mean of all the ranges of each sub group n = number of samples D2 = Statistical Constant used in the estimation of Standard deviation from ranges USL / LSL = Upper and Lower Specification Limits X = mean of n Samples Cpk: A Capability Index, Takes into account process location as well as process variation. Process Variation is derived from the within subgroup range (Sometimes called as short term process capability). Cpk = Minimum [(USL - X) / 3Sigma, (X - LSL) / 3Sigma] Pp = A Process Performance index. Compares the process variation to the maximum allowable variation as indicated by the tolerance. Process variation is derived from process standard deviation (Sometimes referred as overall or long term process capability). Pp does not consider process Location. Pp = USL-LSL / 6 Sigma Where Sigma = Standard deviation Ppk : A Process Performance index, Takes into account process location as well as process variation. Process Variation is derived from the process standard deviation (Sometimes referred as overall or long term process capability) Ppk = Minimum [(USL - X) / 3Sigma, (X - LSL) / 3Sigma] Note : a. if you are planning of performing Cp Cpk or Pp Ppk studies, first ensure that MSA has been performed and your measurement system is reliable. Otherwise what ever you do will be of no use. b. Cpk can never be greater than Cp due to its formulae. In ideal case scenario Cp will be equal to Cpk. c. If sub group sample size is 1 then there will not be cp and Cpk values (eg: continuous produced parts) you will have only Pp and Ppk. In Automobile Industry few customers mandate six pack process capability (refer below image) to accept as process is capable during PPAP stage. Only when 6 pack process capability is met, then approval is provided to go for series production. Final Summary Formulae is same for Cp and Pp, Cpk and Ppk calculation, only difference lies in how standard deviation is calculated. For Cp, Cpk standard deviation is calculated by using within Sub group range. While Pp and Ppk standard deviation is calculated using overall standard deviation. If data is collected at systematic intervals say first 2 parts of every Shift, key dimensions checked in CMM then checking Cp Cpk is preferable. If continuous serialized produced parts (1-30) data is being used then checking Pp, Ppk is preferable. It basically depends on how the data is selected. While doing DMAIC which capability studies can be used. My Opinion Cp Cpk a. If data has been systematically time based picked say once in every hour, 5 parts picked. b. When you have sub group and sample size is >1. c. When using historic data say data captured for first and last part inspection at the beginning of every shift and with that data you want to check process capability. Pp Ppk a. When you do not have historical data all you have is the data that you have produced now. b. When sub group sample size is 1 c. When data is of parts from continuous production (eg: serial 1-50)