Techniques Statistical process control: an essential ingredient for improving service and manufacuring quality

Ben Mason and Jiju Antony

The authors Ben Mason and Jiju Antony are with the Warwick Manufacturing Group, School of Engineering, University of Warwick, UK. Keywords Statistical process control, Service quality, Manufacturing, Quality Abstract Many organizations in the UK are still learning about the implementation and use of total quality management (TQM) techniques such as statistical process control (SPC), even though in other parts of the world, especially in Japan, they have been used widely and successfully for many years. Research carried out by the authors shows that almost the only thing taught to engineers within many UK academic institutions in relation to SPC is control charting. Although the understanding of control charts is a very important aspect of SPC, it is by no means the only one for the successful introduction and development of SPC. It is argued that the emphasis should not be on the control charts themselves, but on the essential ingredients for getting SPC started in organisations and the various issues associated with the implementation of SPC. This paper discusses the ingredients that are needed for the successful implementation of SPC. Electronic access The research register for this journal is available at http://www.mcbup.com/research_registers/ quality.asp The current issue and full text archive of this journal is available at http://www.emerald-library.com
Managing Service Quality Volume 10 . Number 4 . 2000 . pp. 233238 # MCB University Press . ISSN 0960-4529

Introduction
Today's consumer markets experience an ever-increasing demand for better products and services. Customers expect continuously improved quality products and/or services even when they pay less for them than the previous purchase prices. It is therefore logical that if a company wishes to be competitive, one of its main aims should be to focus upon producing products of a consistently high quality. The word ``quality'' can be defined in many ways and has different meanings to different people. However, in order to achieve quality, it is desirable to have a definition for quality that will reflect the true needs and expectations of customers. It is widely accepted that the quality of a product is generally thought of as the ability to fulfil specific needs, or conform to and ideally exceed customer expectations. Quality is therefore an important aspect for any company to maintain competitiveness. During the last half-century, quality and its management has evolved to what is now known as total quality management (TQM). It is probably fair to say that today Dr Deming is recognised as the mentor of what is conceived as TQM. However, there is no standard definition for the term. Existing definitions that are available tend to be very broad. Madu for example defines TQM as:
An organisation-wide quality program to continuously improve products and services delivered to customers by developing supportive organisational culture and implementing statistical and management tools (Madu, 1998).

This definition implies strongly that TQM should be a company-wide policy continuously to improve quality of products and services. Underneath the umbrella of TQM there are several different techniques that can be used to improve product, process and service quality. One of these is statistical process control (SPC).

SPC
SPC is generally accepted to control and manage (management) a process (either manufacturing or service) through the use of statistical methods (Dale, 1994).

SPC is a statistical technique used to control processes and to reduce variation. Variation 233

Statistical process control

Ben Mason and Jiju Antony

Managing Service Quality Volume 10 . Number 4 . 2000 . 233238

reduction is a key aspect to improve quality. There are two main causes of variation, assignable (or special) and common (or chance). Special causes of variation are not inherent in the process and can be therefore readily identified. They are relatively larger in magnitude and require some actions on the process/system to eliminate them. Examples of special causes are resetting of machines, tool wear, errors in measurements, errors in calculations and operator error. Common causes affect all products/services of a process, as they are always inherent in a process. Examples of common causes include humidity fluctuations, temperature changes, electrical fluctuations, deterioration of equipment performance and raw material variations. The main objective of control charts used within SPC is to distinguish between the special and common causes of variation. It is important to note that SPC uses control charts to indicate when there is something wrong with a process and that it is out of statistical control (Caulcutt, 1999). However, it does not tell the user (or operator) exactly what is wrong with the process. Figure 1 shows SPC interactivity. There seems to be a problem with the general thinking behind TQM. In the western world the consensus is that (Oakland, 1999) techniques like SPC should be implemented for customer satisfaction rather than part of a strategic plan by the company. The full benefits of SPC tend to be realised only when the motivation is appropriate:
Organizations that have implemented statistical process control of their own free will experience Figure 1 SPC interactivity

advantages to a greater extent (BrannstromStenberg and Deleryd, 1999).

There are numerous examples indicating the benefits of SPC and why it should be implemented freely as part of an organisation's quality policy. Watson (1998) provides an excellent discussion on this aspect of SPC.

Why do we need SPC?

At the beginning of the century product quality was checked by inspection. In other words, inspection based quality control was used for checking the final product/service quality. However, inspection-based quality control was unreliable, inefficient, costly and time-consuming. Moreover, inspection-based quality control is reactive in the sense that defective items or products will be made before they are found and thus will incur scrapping or reworking costs. Employing people and equipment merely to sort the good ones from bad ones is time-consuming and adds no value in terms of continuous improvement. Inspection does not tell the operators why an error has occurred and what corrective action(s) must be taken to eliminate the error. In order to tackle this problem, a preventive measure must be used at the operation stage in order to ensure that desirable product quality can be achieved. Such an approach based on statistical methods (called SPC today) is used to monitor, control, analyse and improve process performance by systematically eliminating special causes of variation in processes.

234

Statistical process control

Ben Mason and Jiju Antony

Managing Service Quality Volume 10 . Number 4 . 2000 . 233238

.

Figure 2 shows where and how SPC fits into the TQM and competitiveness scenario.

more efficient management, and better understanding of process; reduction in time spent fire-fighting quality problems.

Potential benefits of SPC

One of the main reasons for the failure of SPC implementation within organisations is a lack of understanding the potential benefits of SPC. It is important that SPC awareness and benefits should first be realised by senior management and then be cascaded down through the organisation hierarchy. This should be done before any SPC initiative begins. The following are the typical benefits that can be gained from the application of SPC: . reduction in wasted efforts and costs; . process improvement, greater output; . better consistency of process output; . improved operator information: when to and when not to take action; . a predictable process can be achieved; . a common language on performance of process for different people across departments; . SPC charts help distinguish special from common causes of variation; . variation reduction; . reputation for high quality products/ service and thereby reduce customer complaints; . healthy market share or improved efficiency/effectiveness; . reduced quality costs; . reduced need for checking/inspection/ testing efforts;
Figure 2 Relationship between profitability and SPC

Service quality
Society today is experiencing an everincreasing growth of the service industry. So much so, that around 80 per cent of the GNP can be attributed to the service sector in the USA and many European countries. However, strangely enough, the use of SPC within the service industry is not common. The application of SPC to the service industry can be just as beneficial as it is to the manufacturing industry, in improving service quality and ultimately customer satisfaction. Service quality concerns the difference between the customers' expectations and their perceptions of the service they actually receive from the service provider. Measuring service quality requires an understanding of the service quality characteristics. SPC requires the controllable factors to be identified and the measurable responses to be clearly defined (Xie and Goh, 1999). Lack of understanding is perhaps one of the reasons for the lack of SPC application in the service sector. To improve the quality of a service using SPC, the service performance characteristic associated with the service should be identified and measured. The decision of what and how to measure should be made carefully. The measurement taken should be a fundamental component towards achieving customer satisfaction via service quality. Table I presents a selection of quality characteristics in service industries and their most appropriate control charts. Case studies involving the use of SPC within the service industry have reported the following. Forte plc (Jones and Dent, 1994) carried out a pilot study within the staff restaurant, using SPC to analyse the variety of food, temperature of food and the hygiene in the staff restaurant. In this case Forte used X R charts to monitor all the three variables. The results showed that all three variables had acceptable performance levels, and none was outside the accepted tolerance level. Encouraged by the study, Forte now proposes

235

Statistical process control

Ben Mason and Jiju Antony

Managing Service Quality Volume 10 . Number 4 . 2000 . 233238

Table I Service processes and key quality characteristics in service industries Service Banking Higher education Control chart X R chart Transaction processing time Number of processing errors p chart Satisfaction and fulfilment from obtaining a quality education u chart Proportion of graduates who have job offers within a period of time c chart Number of complaints received by a librarian p chart Proportion of medical errors p chart Number of billing errors per 100 accounts c chart X R chart Time to be admitted in an emergency room Processing time for claims X R chart Number of incorrect entries per 100 documents c chart X R chart Service delivery time X R chart Time to be transferred from the airport to the hotel Number of errors in water/gas meter readings p chart X R chart Waiting time to have a telephone line installed The proportion of electrical power loss in transmission and distribution X R chart Lost/damaged baggage np chart Number of mistakes in reservation c chart X R chart Flight delay in arriving at a destination Quality characteristic

Health-care services

Insurance company Hotel and restaurant industry Public utilities

Transport (airline industry) Source: Mitra (1998)

a more detailed study within a large business hotel. Wood (1994) gives an example of SPC usage within a leisure centre. The time taken to respond to external telephone calls is a key measurement for a leisure centre, because failure to respond quickly may lead to lost business. The leisure centre monitored the number of rings before the phone was answered. The manager decided to use X R charts to try and achieve the ``industry target'' of 98 per cent of calls answered within three rings. The charts showed that at the time 58 per cent of calls were answered within three rings.

Practical difficulties in the implementation of SPC in industries

Dale et al. (1990) conducted a study involving the use of SPC by part/component suppliers in the automotive-related industry. Two surveys were reported. The aim of the first survey was to identify the practical difficulties experienced when introducing SPC. The second showed what the main barriers were when trying to apply SPC and its further development. Out of 158 respondents, 77 per cent indicated that they had experienced difficulties in introducing SPC, and 82 per cent indicated that they had encountered difficulties with its applications and development.

There are many reasons for failure when it comes to implementation of SPC, some of which are mentioned below: . Lack of training and education: this creates problems organisation wide, from the operators to the senior management, because there is a general lack of understanding and awareness of why SPC is being implemented. . Management commitment: management need to commit themselves to SPC, and provide necessary human and economic resources for SPC. . Not understanding fully the potential benefits: this stems from a lack of appreciation of the potential benefits and can result in personnel taking a flippant attitude towards all aspects of SPC. . Failure to interpret control charts and take any necessary actions: if correct training is given to operators then this reason for failure can be eliminated. . Lack of knowledge of which product/process characteristics to monitor and measure: the selection of product characteristics or process parameters are absolutely vital for the success of any SPC initiatives. . Inadequate measuring system in place: there is a great deal of variation in any measurement method and therefore it is essential to ensure that the gauges are capable of doing their intended function.

236

Statistical process control

Ben Mason and Jiju Antony

Managing Service Quality Volume 10 . Number 4 . 2000 . 233238

For a company to be successful with SPC implementation all of these above points should be addressed. It seems possible that the overarching factor incorporating all these problems stems from a lack of education in quality management techniques, SPC in particular. With the correct training and ongoing education, problems such as not understanding the benefits, incorrect interpretation of control charts and incapable measurement system can be eliminated.

Essential ingredients for the successful application of SPC

In order to apply SPC effectively in any organization (either service or manufacturing), it is fundamental to understand the essential ingredients that will make the application successful. There are several prerequisites necessary for successful SPC implementation. Four essential areas that should be the focus are: (1) Management issues total company commitment, creating a responsive environment for actions on the processes/ systems, necessary resources for training and education. (2) Engineering skills understanding benefits, gauge capability, action taken and understanding of processes/systems. (3) Statistical skills statistical stability, choosing, calculating control limits, and interpreting control charts correctly. (4) Teamwork skills company-wide understanding of SPC, its benefits and rewards, requires co-operation and support from all levels of an organization. Figure 3 illustrates the four prerequisites for the successful application of SPC in any organization. First, management commitment is an absolutely essential component for the resolution of common and special causes of variation. Unless top management is prepared to provide the necessary resources and support towards any changes (i.e. in terms of actions on the process/systems), a company is likely to fail in its use of SPC. The next major point for successful SPC introduction is training at all levels, with a bottom-top approach from operators to top management. It is very important to make everyone aware of the key benefits and the role of SPC for continuous improvement of product/process performance. It is important to note that everyone in the SPC team should know why the company is employing SPC for variation reduction in core processes. Furthermore, if sufficient time and resources are spent on training and education, it may be possible to avoid failing in the engineering, statistical and teamwork areas (areas 2-4 above). It is advised to form a process action team (PAT) responsible for taking necessary remedial actions when the process is going out of control. Finally, an SPC implementation

Statistical education for engineers

Perhaps the training and education towards SPC implementation should be taught at an earlier stage and on a wider front in the education system, especially to engineering students in higher education. Currently, within engineering institutions very little time is spent on management and implementation aspects of SPC. The main focus seems to be on control charting of various processes:
Too often organisations look at the ``control chart'' as the only approach to handle issues and this will not work (Xie and Goh, 1999).

It seems that the word ``statistics'' invokes fear amongst industrial engineers (Antony et al., 1998). Very few engineers graduating today from UK higher education institutions are exposed to powerful problem-solving techniques such as SPC, design of experiments (DOE) and Taguchi methods (Antony et al., 1999). This probably explains the problem for employers when trying to recruit graduates with the necessary SPC skills. A user-friendly/practical guide is needed to provide a step-by-step approach of where and how to start implementing SPC in organizations. Rungtusanatham et al. back up this statement:
Very little knowledge has accumulated or has been documented to identify, describe, and define the requisite organisational policies and actions to make the implementation and subsequent practice of SPC an effective and viable part of any organisation's quality management system (Rungtusanatham et al., 1997).

The authors are currently developing a useful and practical framework for the implementation of SPC that will assist industrial engineers with limited skills and knowledge in SPC. The framework will take the form of a systematic methodology for the effective implementation of SPC in any industrial setting.

237

Statistical process control

Ben Mason and Jiju Antony

Managing Service Quality Volume 10 . Number 4 . 2000 . 233238

Figure 3 MEST diagram defining prerequisites for successful SPC implementation

team should be selected with an SPC facilitator to lead and steer the team.

References
Antony, J. and Kaye, M. (1999), Experimental Quality A Strategic Approach to Achieve and Improve Quality, Kluwer Academic Publishers. Antony, J., Kaye, M. and Frangou, A. (1998), ``A strategic methodology to the use of advanced statistical quality improvement techniques'', The TQM Magazine, Vol. 10 No. 3, pp. 169-76. Brannstorm-Stenberg and Deleryd (1999), ``Implementation of statistical process control and process capability studies: requirements or free will?'', Infotrac Electronic Resources. Caulcutt, R. (1999), ``Statistical process control'', Assembly Automation, Vol. 16 No. 4, p. 13. Dale, B. (1994), Managing Quality, Prentice-Hall, UK. Dale, B., Shaw, P. and Owen, M. (1990), ``SPC in the motor industry: an examination of implementation and use'', International Journal of Vehicle Design, Vol. 11 No. 2, pp. 213-18. Jones, P. and Dent, M. (1994), ``Lessons in consistency: SPC in Forte plc'', The TQM Magazine, Vol. 6 No. 1, pp. 18-23. Madu, C. (1998), Handbook of Total Quality Management, Kluwer, London. Mitra, A. (1998), Fundamentals of Quality Control and Improvement, Macmillan Publishers. Oakland, J. (1999), Statistical Process Control, Butterworth-Heinemann, Oxford. Rungtusanatham, M., Anderson, J.C. and Dooley, K.J. (1999), ``Conceptualizing organizational implementation and practice of statistical process control'', Journal of Quality Management, Vol. 2 No. 1, pp. 113-37. Watson, R. (1998), ``Implementing self-managed process improvement teams in a continuous improvement environment'', The TQM Magazine, Vol. 10 No. 4, pp. 246-57. Wood, M. (1994), ``Statistical methods for monitoring service processes'', International Journal of Service Industry Management, Vol. 5 No. 4, pp. 53-68. Xie and Goh (1999), ``Statistical techniques for quality'', The TQM Magazine, Vol. 11 No. 4, pp. 238-41.

Conclusion
This paper attempts to remove the myth that SPC is concerned only with control charting of processes. Furthermore, key factors needed for SPC are defined, namely: . management issues; . engineering skills; . statistical skills; . teamwork skills. Within higher education, there is a need to move away from teaching only control charting of processes, towards a more embracing curriculum regarding SPC. Moreover, engineers should be made fully aware of the necessary ingredients mentioned for successful SPC implementation. A suggestion for the teaching of such quality improvement techniques is to review the successful case studies of SPC implementation. A major issue addressed by this paper is the need for a systematic and practical methodology (i.e. where to start and how to perform an SPC study in an organised manner) for the implementation of SPC in industry. The authors are currently developing such a useful framework which will encourage the wider application of SPC in the UK manufacturing and service organizations with success.

238