APPLYING SPC TO PHARMACEUTICAL COMPANY

APPLYING SPC TO PHARMACEUTICAL COMPANY


Process Capability Index (“PCI”) is defined as ratio of the specification width to the natural tolerance of the process. Cp relates the natural variation of the process with the design specification in a single, qualitative measure,

Cp = UTL – LTL / 6s
Where
UTL = Upper tolerance limit
LTL = Lower tolerance limit
s = Standard deviation of the process

Process Capability Index is used to find out how well the process is centered within the specification limits. It is denoted by Cpk.

Cpk = Cp(1-K)Where,Cp = Process CapabilityK = 2(Design Target - Process Average) / (USL - LSL)Design target is the actual specification targetted without +/- allowance.

Here in this case, the PCI can be used for setting objectives and improving the process.

From the Table 14.5

UCL = 4.96679

LCL = 4.93541

µ = 4.95048

s = 0.008325841

Given the formula

Cp = UTL – LTL / 6s
Cp = 4.96679 - 4.93541 / 6*0.008325841

Cp = 0.62814

Now

Cpu = UTL - µ / 3s = 4.96679 – 4.95048 / 3*0.008325841 = 0.65298

Cpl = µ -LTL / 3s = 4.95048 - 4.93541/ 3*0.008325841 = 0.60336



Histogram:

4.93541
4.93667
4.93901
4.94294
4.94337
4.944487
4.94539
4.94626
4.9484
4.94904
4.9521
4.95252
4.95287
4.95311
4.95385
4.95422
4.95482
4.95533
4.9555
4.95603
4.9571
4.95775
4.95888
4.95941
4.95966
4.96014
4.96016
4.96114
4.96175
4.9623
4.96238
4.96255
4.96543
4.96633
4.96679



Bin
Frequency
Bin
Frequency
4.9350
0
4.9600
8
4.9400
3
4.9550
7
4.9450
3
4.9650
7
4.9500
4
4.9500
4
4.9550
7
4.9400
3
4.9600
8
4.9450
3
4.9650
7
4.9700
3
4.9700
3
4.9350
0





2. The reason why it is incorrect that the operators did not plot the initial data, find special causes, and compute the new control limit is that it led to subsequent modifications and correction in the centering of the process. When the maintenance technician set the length of the adjustment cap to where he was supposed to (lower), the worn-out threads made it impossible to hold the locknut in place, hence resulting in loosening of the locknut and thereby drifting of the center and producing syringes of erratic length. Here, the operators of the first shift who were not trained/familiar in SPC charting used only control limits they obtained to evaluate the future measurements, but did not plot the points. This led to plotting of the control limits with erroneous centricity, hence incorrect observations and adjustments. Therefore, though after the subsequent adjustments by the maintenance technician the syringes produces were of the desired and acceptable length, the variation has reduced dramatically, the string of 15 points of the x-bar chart were all above the center, hence resulting in the process being out of statistical control.
• 3. The learning and observations from this case that can be made out are that being familiarity with the SPC process and charting will be the most important for the process to be successful and quality-oriented. Erroneous data /input collection and analysis could lead to false assumption about the process quality. Also, a careful and out-of-box analysis of the various quality-determining factors such as x-bar chart or R-chart, histograms, SPC, etc. should be done before deriving any conclusion Only, a process that is operating with only chance causes of variation present is said to be in statistical control. It should always be kept in focus that the eventual goal of an SPC is the elimination of variability in the process
•