Beyond Quality in Supply Management
Peter Stannack, Principal Consultant 0+44 1670 815258.
Martyn Osborn, Head of Procurement and Contracting Scottish Hydro Electric plc, Perth United Kingdom 0+44 1738 455420.
82nd Annual International Conference Proceedings - 1997
Introduction. In supply management, we can see both an increasing emphasis upon quality in management and yet an increasing dissatisfaction with it's results. Vendor assessment tools specify 'quality' and 'TQM' as categories for selection, and yet late deliveries and longer lead times still plague companies. Buyers purchase goods from suppliers with quality certification or with TQM programmes and yet products still fail on site or at assembly. Supplier development teams implement SPC training, team working etc., and product reliability does not improve or waste increases dramatically.
This paper will argue that although the quality movement is still alive and well in both Europe and the United States, (Cole 1995), quality is 'dead' for supply management. We will question a number of the claims made for 'Continuous Improvement Firms' (Cole and Mogab 1995), and highlight some of the inconsistencies and internal contradictions of 'quality' for supply management.
This paper will argue that there are fundamental flaws in several areas of quality. The first of these is the inability of managers to differentiate between quality as outcome and quality as process. The second flaw is linked the problem of measurement and therefore communication in quality. The third is the inconsistency between quality as a control mechanism, and quality as a continuous improvement or learning mechanism. The paper will examine each of these flaws briefly, before considering why purchasing may have taken so strongly to 'quality' and what purchasers might need to do to go "beyond quality" in supply management.
Considering Quality. Quality is one of the watchwords of purchasing. Quality is one of Bailly and Farmers 'Five Rights'. Few authors, however, offer a consideration of quality in purchasing for exceptions (McRobb 1988, Cali 1994). We do, of course face problems when we talk of quality generically. Engineering quality and service quality are often perceived as different elements of 'quality'. Nonetheless, a wide number of generic definitions exist - "fitness for purpose", "continuous improvements" etc. (see Exhibit One)
As noted above, the first issue which faces the purchaser in managing supply is that of specification. The widening focus of the purchasing profession has been noted elsewhere. Focal shifts have meant that purchasers have become strategic supply managers; purchasing has broadened to concern itself with general supplier performance and even 'relationships' (Hanmer Lloyd 1996) As this focal shift has taken place, purchasing as supply management has become problematic. A broadening of focus implies a broadening of analysis; a broadening of analysis implies a broadening of intervention. Purchasing as supply management involves itself in all sorts of process issues including supplier quality, supplier development and supply networking.
Despite a growing concern for process we can say that purchasers should 'buy' only outcomes i.e. product consistency, product reliability, product cost. Although they often need to pay attention to process i.e. the manner in which these outcomes is delivered, there is no value to be acquired in process, value can only be acquired through outcomes. Quality as outcomes is, to use Crosby's phrase, free. Quality as process can be extraordinarily expensive. Some indicators of this expense can be found in Chrysler, Ford and General Motors QS9000 programme. QS9000 is a strengthened, adapted version of ISO9000 and each of the "Big Three" have made a commitment to implement QS9000 in all first time suppliers by 1998. In the interpretations of QS9000, published bi-annually, it is recommended that up to twenty one person days are spent on QS9000 audit, with five and a half of these days spent on site. This audit period does not include pre-assessment, pre-audit document review, corrective action and registration. Audit for up to 500 suppliers could take between three and ten thousand person days - fifty WTE posts.
Exhibit 1 - Definitions of Quality
Deming - "Quality is a predictable degree of uniformity and dependability at low cost and suited to the market.
Juran - "Quality is fitness for use"
Crosby - "Quality is conformance to Requirements"
Taguchi - "Quality is the (minimum) loss imparted by the product to society from the time the product is shipped"
Hoshin - "Quality is correcting and prevention loss, not living with loss"
British Standard Definition - "Quality is the totality of features and characteristics of a product, service or process which bear on its ability to satisfy a given need; from the customers viewpoint"
Feigenbaum - "Quality is in it's essence a way of managing the organisation"
Admittedly, the Big Three do carry out audit through third parties who charge for the task. Nonetheless, this wastes resource somewhere within the value chain, when the whole focus of supply management has been to drive cost out of the 'chain'.
This leaves us with a problem when purchasing 'quality'. Is quality a process or an outcome. The literature is far from clear. Some definitions of quality are based around outcomes but others are based around processes. These two categories are, of course, mutually exclusive. An outcome is a target or state to be reached, a process is a path or mechanism for reaching that state or target. Quality cannot be both a road and a destination, unless it is similar to the 'wavicle' in quantum physics which can act as both a particle and a wave in certain circumstances.
Supply management is not, however, concerned only with explaining but in predicting and controlling supply. The 'road' metaphor is useful when managing quality within organisations. It is less useful when buying quality from organisations. The purchaser may wish to pay someone to travel to a particular destination. To be told that the individual is always travelling hopefully, never to arrive is somewhat disappointing. Ambiguity, (Sabini and Silver 1982)is a resource. Ambiguity of purpose within organisations allows managers room for manoeuvre - deniability or claims can be distributed according to preference in ambiguous situations. Nonetheless, ambiguity between organisations can lead to another set of problems.
This is perhaps best explained by the work of the logician WRO Quine and takes us into the realm of language and what might be termed "leakages of meaning"
Quine (1960) asks us to imagine a linguist studying a newly discovered tribe. A rabbit scampers past and the native shouts "Gavagai" What does "gavagai" mean. It could mean rabbit, it could also mean a particular rabbit, a small furry mammal, a particular type of rabbit, a scampering thing the verb "to scamper". In fact gavagai could mean (almost) anything. Similarly the definitions offered in Exhibit One show problems with purchasing something other than a 'quality rabbit' Other titles within the literature also show potential problems with the introduction of quality programmes. Texts such as "Managing Quality for Higher profits" and "Profit by Quality" seem to demonstrate one type of motivation
This is the problem which purchasers face. How do they specify the type of quality they require?
Specifying Quality. We do in fact overcome the "gavagai" problem by specifying quality numerically. Quality is, for instance expressed in measurable items such as "zero defects" or "zero range gaps". In purchasing this means actually defining quality as consistency of size, weight, etc. or as availability to demand. There are no reasons why we should label these specifications as quality and positive drawbacks to so doing. We can, however consider some other problems with specifying quality.
Quality has, as we can see, become almost an ideology or belief system (Tuchman 1995) As such many characteristics are associated with the term. One of the more salient of these characteristics is that of customer focus or customer centredness. Indeed, a whole subset of the general quality movement - QFD or quality function deployment is associated with customer focus. Some conceptualisations of quality, in fact seem to rest upon this customer responsive dimension. For some authors, "Total Quality" is the consistency /reliability construct of quality with a customer focus added.
As soon as we add a customer focus dimension to quality however, we can see that things become difficult. In the field of service quality Johnson (1995) has carried out an integrative review of the determinants of service quality in which he identifies either process characteristics such as helpfulness, care, courtesy, communication.
These determinants are clearly vital to the delivery of quality but so many are abstract concepts without clear measures. Other problems exist in determining just how satisfying or dissatisfying these determinants are. Moxey and Sanford (1993) have noted the effects of context and focus in the communications of quantifiers.
Here individuals do not accurately communicate quantitative information when using non metric terms. Their research showed clear differences in interpreting terms such as a "very good" service or they "rarely" let you down, "Some" of the staff are excellent. These differences in estimation were as high as 30% in one study. To paraphrase Lord Kelvin, we can see that what cannot be measured cannot be purchased effectively.
In the last two sections we considered the problems where quality was seen as an outcome by purchasers and a process by suppliers. We also considered the problems with actually measuring, and therefore purchasing quality. In the next section we will consider another issue which springs from the consideration of quality as a control mechanism and quality as a learning or development mechanism.
Quality as Control and Quality as Learning. Earlier in the paper we looked at the blurred line between quality as process and quality as outcome. The initial results of our research show that quality is perceived as a process by quality managers and outcomes as purchasers. One relatively easy way of overcoming the problem of differentiation is to represent quality as a set of targets - zero defects.
Such target setting mechanisms are, of course, problematic. In the field of finance, Goodhart's law states that "any measure which is used as a target becomes a bad measure". There is no reason to suppose that this law is not applicable in other circumstances. We can see examples in shop floor productivity where staff who "soldier" to reduce production targets are met by manager's introduction of slack adjusted norms. The performance of middle management could be assessed through return on investment Managers then fail to invest to ensure good ROI Senior management then introduce retained profit as a target/measure.
This escalation of measurement and systems demonstrates the pitfalls for supply management. As financial systems become less adequate to drive performance (Nickell 1995, Proams 1996) so qualitative measures are used more extensively.
Nonetheless, such target setting mechanisms are prevalent in a number of settings (Unipart 1995) We would argue that it is possible to identify within quality managers a range of different cognitive orientations about quality. These orientations will effect attitudes and beliefs with regard to the objectives or functions of quality. They will also effect the way in which quality programmes are implemented. This orientation is multi dimensional (Stannack 1996) and includes constructs such as control, improvement, waste elimination etc. Two of these constructs are of particular concern to the supply manager.
The first of these is the prescriptive construct which is linked to control and management. Here we can identify a 'pokayoke mentality' where the process becomes more important than the outcome. Constructs such as this can be identified in work by Greene (1993) who seems to elide 'lean production' with quality. He identifies different types of 'fat' within the organisation including 'opinion' which is 'negated by quality' (p280) in Greene's version of 'global quality' all potential or actual variance is eliminated by 'quality'.
The second is the responsive or customer centred construct. This is often subordinated to the control construct because the opinion held by many practitioners is that 'if you cannot control your company, you cannot supply customers'. Examples of the responsive construct can be identified within the literature in work by authorities such as Shiba et.al (1993) who see quality as being "Fitness to latent requirements" This goes beyond Juran's 'fitness for purpose' and into the realms of prediction or creation of customer need such as that practised by Sony in it's marketing of the Walkman.
There will, of course be issues with regard to the degree to which an organisation can 'walk the tightrope' between responsiveness and prescription. Many factors will impact upon this walk but it is possible, by extension to identify different types of quality in supply ranging from responsive quality to prescriptive quality. Supply managers should, one would expect, be interested in responsive rather than prescriptive quality.
We can see that if we were to use these dimensions to describe a grid, high prescriptive and high responsive would be the ideal for purchasing. This would be the least expensive type of supply management. Low prescriptive and high responsiveness would require more management of supply. Low prescriptive and low responsive quality firms should be avoided and high prescriptive, low responsive firms will be suitable for non agile contracts.
In the final section of this paper we will outline the way in which one company has eliminated ambiguity from its supply management strategy and set out clear "rules" for the "game" which suppliers can respond to effectively.
Changing the Rules of the Game. We saw above that ambiguity can be a resource. The reasons for this ambiguity have been explored elsewhere (Stannack and Scheuing 1997). Scottish Hydro-Electric plc (H-E) is an electricity generating company based in Scotland which exports over half it's product to the rest of the United Kingdom. H-E face unique geographical and supply base management challenges due to the varied nature of their supply needs. Recently H-E have been working to improve supplier performance by creating a set of standards for general performance. These standards cover outcome dimensions such as timeliness, cost, delivery accuracy, etc. They also cover process dimensions such as communication, planning, measurement, etc.
In developing these standards, H-E wanted to avoid a number of the pitfalls which they saw in other supply management systems. The Procurement and Contracting department (P&C)had reviewed a number of systems and saw that quality was being increasingly used as a change management tool by purchasers or supply development teams, particularly within the automotive assembly industry. Given the reasonable effectiveness of TQM as such a tool, this seemed reasonable. It offered suppliers a symbol which they could follow, and one which was not necessarily closely linked with one particular customer. This made it easier to achieve take up from suppliers.
P&C believed, however, that the ultimate objective of TQM programmes in suppliers was not recognised, even by those purchasers employing the technique. Given the difficulties of predicting supplier performance effectively, purchasers were using TQM to 'drive out opinion fat' and create suppliers who were as much like themselves as possible. This is one variant of the strategy used in Japanese dependency sourcing, where supplier relations are based upon Amae or obligation/dependency, and consequently innovation is driven out of the supply network.
H-E, however, saw innovation in suppliers as being a major determinant of competitive advantage. In order to encourage responsiveness in supply H-E split their performance standards into two categories, each of which was linked loosely to processes. The first set of standards were 'hard' standards which could be both measured and specified. The second were 'soft' standards which could be measured but not specified. By linking these standards to accurate measures which were not set as targets, H-E has seen a dramatic improvement in the way in which it works with suppliers and a dramatic improvement in supplier responsiveness and performance.
A Nelley, M Gregory and K Platts (1995)"Performance Measurement System Design: A Literature Review And Research Agenda." IJOPMM Vol 15 No 4 1995 pp 80-116
P Dahlen, J Ericsson and H Fujii (1995)"Labour Stability And Flexibility Conditions To Reach Just In Time" IJOPM vol 15, no 9(1995) pp 26-44
W Qualls and J A Rosa (1995)"Assessing Industrial Buyers Perceptions Of Quality And Their Effects On Satisfaction" Industrial Marketing Management 24 259-368
W.E. Cole and J W Mogab (1995) "The Economics of Total Quality Management" Blackwell Ltd. Oxford
S.Shiba, A Graham, D Walden (1993)"The New American TQM", Productivity Press, Portland, Oregon.
R.T. Green (1993) "Global Quality" ASQC Quality Press Business One Irwin 1993
R. Hale, R. Howal, D. Carlton and T. Sehnert "Managing Supplier Quality" Monochrome Press, Exeter NH
McRobb, M (1989) "Purchasing and Quality" Marcel Dekker New York and Basle
Crosby P.B (1979) "Quality is Free" Mentor New York
Deming EW (1982) "Quality, Productivity and Competitive Position" MIT MA
Feigenbaum A V (1986) "Total Quality Control" McGraw Hill, New York
Ishikawa K (1976) "Guide to Quality Contgrol" Asian Productivity Organisation Tokyo
Juran J M (1988) "Quality Control Handbook" McGraw Hill, New York
Taguchi G (1985) "What is Total Quality Control - The Japanese Way" Prentice Hall Eaglewood Cliffs NJ
Shingo S (1986) "Zero Quality Control. Source Inspection and the Poka Yoke System" Productivity Press Cambridge MA
Hanmer Lloyd SA (1996) "Relationship Appraisal - A Route to Improved Reseller Challenge Performance Industrial Marketing Management" 25 173-185
Sabini J and Silver M (1982) "The Moralities of Everyday Life" Oxford University Press