The Certified Quality Engineer Handbook Examples from Ch

The Certified Quality Engineer Handbook Examples from Ch. 37: Statistical Process Control (SPC) Dr. Joan Burtner Certified Quality Engineer Associate Professor of Industrial Engineering and Industrial Management

Quality as a System Characteristic Goal: A stable Quality Management System (QMS) with well-defined policies and procedures and a culture of compliance with those procedures. System Characteristics: System Out of Control One or more processes exhibit special causes. System in Control There is no evidence of special causes within the system. Improved System in Control Stable processes are continuously being improved. Source : Dean (2013) Lean Healthcare Deployment and Sustainability ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 2

Borror: Chapter 37 Review of Charts q Construction of Control Charts Ø Control Charts for Attributes Ø Control Charts for Variables q Interpretation of Control Charts Ø “Manual” Application of Tests Ø Statistical Software Application of Tests q Other Process Charts ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 3

Control Limits, Random and Nonrandom Sample Observations Upper Control Limit (UCL) Non-random 99. 7% +3σ Process Mean Lower Control Limit (LCL) -3σ 1 2 3 4 5 6 7 8 9 10 11 12 Sample number Source: Ozcan Figure 12. 4 (Modified for Three Sigma Limits) ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 4

Statistical Control Chart Types Attributes c-chart p-chart Variables(Subgroups) u-chart Mean Charts (X-bar Charts) σ Method ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Variation Charts Range Method Slide 5

Variables Control Charts (Continuous Data) When process characteristics can be measured, variables control charts are the appropriate way to display the process monitoring. The Xbar-chart and the Range chart are displayed and interpreted together. When the Range chart exhibits out-of-control status, the rules for evaluating the Xbar-chart should not be used. The Xbar chart is appropriately evaluated after the Range chart indicates that the process is in control. Use caution in interpretation when using statistical software that evaluates both charts simultaneously. See examples on pages 496 -499. ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 6

Variables Control Chart for n = 1 Variables(Subgroups) Variables (Individuals) Individual observation Moving Range Mean Charts (X-bar Charts) Note that the tests that apply to the subgroup charts do not apply to the Individuals Charts. ISE 428 ETM 591 JMB CH 37 Examples σ Method Spring 2014 Variation Charts Range Method Slide 7

Attribute Control Charts (Discrete Data) When process characteristics can be counted, attributebased control charts are the appropriate way to display the process monitoring. The p-chart is the appropriate control chart for a process with only two outcomes (defective or not defective) when the proportion defective is calculated. The c-chart is the appropriate tool to display monitoring if the number of occurrences per sampling period is recorded. The u-chart is the appropriate control chart if the number of occurrences and the number of items per sampling period is recorded. The average number of occurrences per sample is calculated. ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 8

Attribute Control Charts (Discrete Data) See text for examples of p-chart. See text for examples of c-chart. We will discuss the u-chart example in class. ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 9

Borror: Example 37. 4 A random sample of 1000 account activities is collected weekly for 12 weeks. The institution would like to monitor the proportion of errors. Week Activities in Error Week 1 6 7 9 2 11 8 8 3 4 9 12 4 10 10 7 5 5 11 12 6 30 12 11 ISE 428 ETM 591 JMB CH 37 Examples Activities in Error Spring 2014 Slide 10

37. 4 Chart Construction and Analysis Are the data discrete or continuous? Discrete Do the data fit the requirements of a p , c, or u chart? p chart What tests for special causes apply to this type of chart? Extreme point, trend, shift in mean, oscillation Zone rules (A, B, C) do not apply to p charts ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 11

37. 4 Chart Analysis Original control chart produced using Minitab 15 Evidence of special cause (extreme point) ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 12

37. 4 Process Improvement Assignable cause is identified for the out of control point. Sample point 6 is eliminated and a new control chart is produced using Minitab 15. ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 13

Other Charts Cumulative Sum Charts EWMA Charts Moving Average Charts *******Pre-control Charts ******* ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 14

References / Contact Information Borror, C, Ed. (2009). The Certified Quality Engineer Handbook 3 rd edition, Milwaukee, WI: ASQ Quality Press. Dean, M. L. (2013). Lean Healthcare Deployment and Sustainability. New York, NY: Mc. Graw Hill. Ozcan, Y. (2009). Quantitative Methods in Health Care Management 2 nd ed, San Francisco, CA: Jossey-Bass. Contact Information: Dr. Joan Burtner_J@Mercer. edu ISE 428 ETM 591 JMB CH 37 Examples Spring 2014 Slide 15