Training for Quality Simulation for training in quality control Jim Freeman Nikolaos Evangeliou
Article information: To cite this document: Jim Freeman Nikolaos Evangeliou, (1996),"Simulation for training in quality control", Training for Quality, Vol. 4 Iss 1 pp. 27 31 Permanent link to this document: http://dx.doi.org/10.1108/09684879610112837 Downloaded on: 29 April 2015, At: 05:20 (PT) References: this document contains references to 10 other documents. To copy this document:
[email protected] The fulltext of this document has been downloaded 1439 times since 2006*
Downloaded by UNIVERSITY OF MANCHESTER At 05:20 29 April 2015 (PT)
Users who downloaded this article also downloaded: Hans van der Bij, Jeroen H.W. van Ekert, (1999),"Interaction between production control and quality control", International Journal of Operations & Production Management, Vol. 19 Iss 7 pp. 674-690 http:// dx.doi.org/10.1108/01443579910271665 Naceur Jabnoun, (2002),"Control processes for total quality management and quality assurance", Work Study, Vol. 51 Iss 4 pp. 182-190 http://dx.doi.org/10.1108/00438020210430733 Sohail S. Chaudhry, Nabil A. Tamimi, John Betton, (1997),"The management and control of quality in a process industry", International Journal of Quality & Reliability Management, Vol. 14 Iss 6 pp. 575-581 http:// dx.doi.org/10.1108/02656719710186191
Access to this document was granted through an Emerald subscription provided by 320482 []
For Authors If you would like to write for this, or any other Emerald publication, then please use our Emerald for Authors service information about how to choose which publication to write for and submission guidelines are available for all. Please visit www.emeraldinsight.com/authors for more information.
About Emerald www.emeraldinsight.com Emerald is a global publisher linking research and practice to the benefit of society. The company manages a portfolio of more than 290 journals and over 2,350 books and book series volumes, as well as providing an extensive range of online products and additional customer resources and services. Emerald is both COUNTER 4 and TRANSFER compliant. The organization is a partner of the Committee on Publication Ethics (COPE) and also works with Portico and the LOCKSS initiative for digital archive preservation. *Related content and download information correct at time of download.
Introduction
Simulation for training in quality control
Statistical quality control (SQC) methods are of immense importance in the (unceasing) quest for quality improvement. Control charts – originated by Shewhart[1] at Bell Laboratories in the late 1920s – are probably the foremost SQC tool in use today. In a typical control chart – see Figure 1[2] – observations of a particular process characteristic (e.g. a product weight) are plotted against time. The centre line (CL) of Figure 1 represents the target or standard value for the characteristic against which observations are checked. (Note that observations may be based on individual items – or more likely, random samples of items – from the process.) The UCL and LCL lines represent the upper and lower control limits, placed respectively three standard deviations above and below the centre line. If plotted points fall between the UCL and LCL lines and are scattered at random about the CL the inference is that the process is “in control” (Caulcutt[3]). Otherwise it is deemed to be “out of control” – in which case efforts are made to identify possible underlying (“assignable”) causes in order that appropriate action can be taken.
Downloaded by UNIVERSITY OF MANCHESTER At 05:20 29 April 2015 (PT)
Jim Freeman and Nikolaos Evangeliou
The authors Jim Freeman is a Lecturer and Nikolaos Evangeliou is an MSc student both at Manchester School of Management, UMIST, Manchester, UK. Abstract Demonstrates the training value of a new computer-aided learning package, SQCC_ATT. Control charts are used extensively in quality management, and computer simulation provides a valuable facility for generating charts under a variety of different working assumptions. Warns about the potential for misuse of control charts by untrained and inexperienced staff. Concludes that specialist simulation games are likely to play an increasingly important role in training in management control techniques.
Types of data The chart in Figure 1 is particularly suited to handling variable (continuous or measured) data. The latter can refer either to process parameters (e.g. temperature, pressure) or product characteristics (e.g. length, weight). In contrast, attribute (Bernoulli) data are binary-valued: for example, items may be classified as good (pass) or bad (fail ) or lots of items as acceptable (go) or unacceptable (nogo). Countable data relate specifically to the number of defects per item. Each of these different data types gives rise to its own individual class of control charts[4]. Our particular concern is with charts for attribute data and the paper details a new computer-aided learning (CAL) package, SQCC_ATT, for simulating such charts. The SQCC_ATT package complements simulation gaming software produced previously[5] for handling variable data[6].
Training for Quality Volume 4 · Number 1 · 1996 · pp. 27–31 © MCB University Press · ISSN 0968-4875
27
Simulation for training in quality control
Training for Quality
Jim Freeman and Nikolaos Evangeliou
Volume 4 · Number 1 · 1996 · 27–31
(1) np charts which cover the number; (2) p charts which cover the proportion of non-conforming units from a process.
Figure 1 A typical control chart Observation value X
Here n refers to the sample size on which observations are based and p to the probability of a defective item occurring. Note that for p charts, n does not have to be fixed (the normal convention for np charts).
UCL
X X
X
X
CL
X
X
‘…By experimentation e.g. introducing different shifts in the process mean as charts are being screened, the sensitivity of schemes to changes can be observed. In this way, users can swiftly learn to discriminate between the different schemes in different circumstances…’
X LCL
Downloaded by UNIVERSITY OF MANCHESTER At 05:20 29 April 2015 (PT)
Time
Attribute control charts Attribute charts are generally less favoured to variable charts because of their relative insensitivity. Overall, however, attribute charts are simpler, cheaper and quicker to operate and are probably the only practical option where items need to be checked against a number of different benchmarks[7]. Various sampling options exist for handling attribute data: • single sampling; • double sampling; • multiple sampling; and • sequential sampling[2].
A useful technical summary of these and other attribute control schemes appears in John[8].
SQCC_ATT SQCC_ATT is written in TURBO PASCAL (Version 6) and runs on an IBM PC computer or compatible. The package is menu-driven and functions at two basic levels: (1) demonstration; and (2) testing. Both of these rely on the package’s capability for simulating commonly-used attribute control charts.
SQCC_ATT is concerned only with single sampling schemes and in particular with the two basic categories: Figure 2 Attribute control charts: SQCC_ATT options
SQCC_ATT Control charts for controlling the process number non- conforming, (np)
[1]
Direct comparison of two of the following control Shewhart, Modified Shewhart, CuScores, CuSum
