Introduction- Six Sigma Background and Issues
Six Sigma is the practical application of a theoretical statistical measurement that equates to 3.4 defects per million opportunities -a position of practically zero defects for any process or service. Its attainment is one of the highest measures of quality and is based on the ideology that practically all errors are preventable (Behara et al, 1994). Initially originating in Motorola Inc. in 1985 as a response to drastic quality improvement pressures from the threat of Japanese competition (Harry & Schroeder, 2000), it quickly gained many followers particularly G.E., Allied Signal, Ford Motor Company etc. and more recently attentions have shifted to service environments.
Bob Galvin former CEO of Motorola stated that the lack of initial investment in the non manufacturing areas of the business over four years was a blunder that cost the business over 5 million dollars (Basu & Write, 2003, p43). However, organisations have implemented six sigma initiatives in transactional frameworks with success- testimonial for six sigma triumphs in services range from American Express and PriceWaterHouseCoopers to local NHS departments.
The nature of Six Sigma and it’s Quality Objectives
As outlined in Lagrosen & Lagrosen (2003) six core principles form the basis of quality management, constitute the common material measured by numerous recognised quality awards (e.g. Malcolm Baldridge Quality Award, Swedish Quality Award etc) and form the basis of ideas presented by leading authors in this field (e.g. Dale, 1999, Bank, 2000 etc). These six core values are –
1. Customer Orientation
2. Leadership Commitment
3. Participation of Everybody
4. Continuous improvements
5. Management by Facts
6. Process Orientation
Six Sigma methodologies encompass all of these areas and thus in a sense is not revolutionary, rather it’s focus on directing resources and effort towards explicit goals with concrete objectives using a prescribed approach makes it unequivocal and robust to implement in organisations. Goal setting research indicates that there is a strong positive relationship between setting challenging, measurable, specific goals and performance (White & Locke, 1981). Linderman et al (2001) argues this is one of Six Sigma’s foundations of success. Thus Six Sigma may be succeeding in a manner TQM could not– TQM was often criticised for being weak – “It is very difficult to motivate and justify what seems to be a repeated circular path, where what in fact is required is a spiralling process that moves forward with each revolution” states Tennant (2001, p 35) in regards to the unclear targets of TQM.
This common goal in Six Sigma organisations is to reduce costs by eliminating defects (Greatbanks, Lecture- 18/11/03).
Costs of Defects
It is argued that Six Sigma should be implemented through the processes that affect customer satisfaction and organisational effectiveness to reduce costs (Eckes, 2003, p3). The following costs are associated in services:
· Verifyable Failure costs- service defect is detected by customer and brought to the attention of the server for rectification, e.g. a hair is found in the soup at a restaurant, the soup must be replaced.
· Nonverifyable Failure costs- difficult to measure ‘hidden’ costs that are not reported back, e.g. people rarely complain and ask for a refund if they attend a bad theatre production.
-Issues include declining image and goodwill due to negative word of mouth and the costs associated with regaining a lost customer (3-5 times more expensive than attracting a new one) Without a loyal customer base a service organisation would be financially very unstable.
· Internal Failure Costs- costs of correcting defects uncovered by the producer before they reach the customer e.g. Slightly overbooking for an excursion means the service provider needs to book 2 minibuses instead of one.
-Often internal failures result in higher staff turnover and lower morale which in turn leads to recruitment and training costs above the overt costs of rectifying the problem.
(Heskett et al, 1990, p76)
The Costs of Poor Quality (COPQ) corresponds with sigma levels, for instance if Six Sigma has been attained, the COPQ is less that 1% of cost of sales, while operating at a three sigma level, which many companies do, equates to a COPQ level of approximately 25-30% of cost of sales (Basu & Wright, 2002, p39). This demonstrates what a powerful tool Six Sigma can be in reducing costs.
Six Sigma is very relevant for services as it has been found that the costs of quality in service organisations are greater relative to manufacturing (Asher, 1987)
The Nature of Services
Services are notorious for their wastage, inefficiencies and variability (George, 2003, p3), and as the basis of service is human delivery, one may assume that clear goals and a prescribed system of change could motivate transformations in the workforce. However there are more issues that have their roots in the nature of services that effect how Six Sigma can be implemented in such a context.
Six Sigma was initially designed within the framework of the manufacturing company. It is important to note that services differ in nature to physical products in the following regards:
Inseparability – The customer is involved in the actual production process- the service is delivered and consumed at the same time.
Perishability – Being intangible, the service cannot be stored.
Heterogeneity – difficulties in standardising services every time for every customer.
(Ghobadian et al, 1994)
Quality is an important issue in services due to the features of inseparability, intangibility and perishability. That which can not be stored and is intangible cannot be checked for defects before ‘delivery’ to customers. In addition each individual involved in the exchange process brings with them varying levels of expectations and levels of satisfaction in addition to the unpredictable nature of human beings. It is this dominant role of human interaction in services that shape customers expectations and create difficulties in understanding and implementing quality initiatives (Behara & Gundersen (2001)).
The most commonly used definition of quality is the extent to which goods or services meet or exceed customer expectations (Zeithaml, 1981). Customer satisfaction should lead to repeat utilisation of the service; so if ‘zero defects’ are the goals of manufacturing then ‘zero defection’ should be the sign of quality coming to services (Reichheld & Sasser, 1990). Thus for the Define stage of the Six Sigma methodology the areas linked to optimising customer satisfaction should be concentrated upon. Yet it is important to stress that this in itself can be a muddled and complicated feat.
Six Sigma strives for Total Customer Satisfaction in services (Erwin, 2000).
As illustrated by Behara et al (1994, p12) customer satisfaction is a multistage process where levels of satisfaction are multiplied as different facets of the service are exposed to the customer. These facets cover a broad range from ethical practices of the business to timely response to knowledgably staff etc. So for instance no matter how fresh and tasty a McDonald’s burger is, for a customer who has moral issues with the low wages of their employees, fulfilment will never be attained. The key notion is that different customers have different patterns of expectations for the components involved and so, is it possible to have zero customer defection? Not everyone likes the same things and thinks in the same way and thus the service provider must focus on the elements that will please the majority only.
Also as services are intangible, there are greater problems in the measurement of quality, as discussed, what constitutes quality may be different for different individuals based on their perceptions and past experiences and thus what defines defect in services? Often this will be an obvious matter of simply delivering what is promised, yet in most cases reliance on customer feedback, complaints and measurement (as demonstrated in the case study) will have to be used for enlightenment of issues. Six Sigma advocated the measurement of such variables as the only way to gain insight into service defects.
Implications for Services
The use of quality programs in relatively high in the service environment, for example Robinson (2003) found that 90% of the sport and leisure facilities managed by local authorities implement some quality scheme, however it follows that the type of quality schemes in services are considerably less ‘technical’ based (e.g. Statistical Process Control, Design of Experiments, Quality Circles and Failure Mode and Effects Analysis- FMEA) than those found in manufacturing and more in tune with ‘softer’ cultural issues and creating an proficient and efficient climate through employees, not processes (Lagrosen & Lagrosen, 2003). But as Tennant (2001, p36) puts so eloquently this is not the purpose of Six Sigma- “Six Sigma has the tools and the power to cut ice where hot air has contributed little in the past”.
The Six Sigma methodology relies heavily on statistical analysis; traditionally services have minimal data and examination of their techniques, thus this may poise an initial hurdle. Over and above many individuals have a fear of metrics and don’t connect their use to services. Breyfogle (cited from Smith, www.qualitydigest.com) explains “They (services) don’t appreciate the importance of creating meaningful metrics that give insight into how their business processes perform over time. This can lead to fire fighting common cause variability as though they were special cause”. He argues that only the use of statistical control charts will enable services to focus on prevention rather than reacting to problems. Monitoring processes is the only way to progress from subjective hypothesising of reasons of error to concrete data and this one of the main principles of Six Sigma.
Is this fear of metrics justified? Many academics have confronted the problem of applying statistical techniques to non-manufacturing environments, for example Deming (1987, Ch7) gives a long listing of measurements in service industries where SPC or similar can be applied. It is noted by Oakland (1989, p226) that “Data is Data…whether numbers represent defects or invoice errors, the information relates to machine settings, process variables, prices, quantities, discounts, customers, or supply points is irrelevant, the techniques can always be used”. The inference is that statistics can be transferred to services; it is rare though, that problems and issues are documented in the literature (merely success stories) this does not mean however by implication these problems do not exist (Wood, 1994). It will often involve creativity and flair to apply statistical techniques to services in a fashion that causes true understanding of what the reality is through numerical representation.
Healthcare Case Study
(Kooy et al, 2002)
The following example is simplified and divided into the common methodology of DMAIC to illustrate how Six Sigma is implemented in services.
In June 2001 VirtuaHealth, an organisation of 4 Hospitals in New Jersey USA, put together a team of internal employees including frontline staff members and a six sigma project leader (black belt trained by GE Medical Systems), the aim being to –Ensure safe and effective acute anticoagulation capability.
The project would focus on the drug Heparin (an anticoagulant) which was used for the treatment and prevention of thromboembolic diseases (blockages in the veins). Patients administered Heparin within 24 hrs of detection of problems saw a significant reduction in future problems, but there were side affects involved also with the Heparin therapy, including serious bleeding (thromboctopenia) and life or limb threatening thromboses. As a result, a weight based protocol was used to administer correct dosage of the drug.
Team identifies customer (i.e. patient) requirements and process deliverables as preventing or addressing anaemia and thromboctopenia during therapy.
These 2 attributes are defined as follows-
Anaemia- drop in haemoglobin at rate of at least 1g/dL per day (and final value less that 12g/dL)
Thrombocytopenia- 50% drop in platelet count (enables blood clot) or a count less than 100,000.
– Occurrence of these attributes was considered out of the ‘Therapeutic Range’.
Acceptable practice was defined as the recognition of the reduction at monitoring stage and actions being taken by the physician- discontinue heparin therapy or other such appropriate measures.
Team used pharmacy and laboratory databases and manual data to measure current performance. From the 815 patients who had received therapeutic doses over the last 6 months, 18% were sub-therapeutic and 35% were supra-therapeutic.
The Team constructed a high level process map to better understand the flow of activities involved in administering and monitoring Heparin (this is the service component).
The mean time from administering the drug to monitoring the outcomes was 8.5hrs, which was considered late but acceptable, yet it was the amount of variation in mean times that was causing problems. Samples being drawn early could lead to drug adjustments based on non-steady state results, whilst those drawn too late could result in an unacceptable diffusion rate of the drug being administered.
This phase entails the identification of the factors that drive the process results. Barriers towards successful completion of each process step was identified and a more detailed process map was drawn (including the laboratory and pharmacy sub cycles), a total of 92 steps were identified for reaching completion of first dose adjustment.
Many problems were identified. Step completion was often down to staff remembering to act, often hours after triggering the event. It was concluded that the complexity of the system was impeding performance and there were few system elements in place to help prevent problems. In particular, the initial step using the weight based protocol was rarely followed meticulously due to time constraints- only 48% of patients were being weighed at all (critical for accurate measurement of drugs) and out of the remaining patients where the weight was estimated, 20% of estimates were more that 10% off. Finally the progression from each step was disjointed and there was often uncertainty as to who had responsibility for the various stages.
In summary, adverse outcomes were not due to minor process variation; rather, they were connected with major break-downs in the delivery of procedure. The team concluded that by simplifying the acute anticoagulation method and error- proofing each stage this would act as the greatest prospect for ensuing safety.
At this juncture, the implementation and measurement of changes to the process toward desired performance is considered. The weighing problem was overcome by investing in beds that had integrated scales, which the hospital used in other departments with much success for routine weighing. This problem had been “flying under the radar” for several years and had only been made explicit through the Six Sigma intervention. This is coupled with an administration record for the weight based Heparin protocol that notes the responsibility (given by doctor taking on case to shift nurse) of re measurement in the agreed time of 6 hours. In addition, new infusion machines that restricted the range of infusion based on the weight calculation were implemented to reduce the possibility of overdose due to lapse of nurse attention.
Visible metric or ‘dashboards’ (control charts, run charts, reports etc.) are used by the project owner to ensure performance is sustained at optimal levels. The performance will also be tracked on a monthly basis by a local quality analyst in the hospitals quality assurance department. Deviations are to be reported to and reviewed in detail by the quality director and pharmacy and therapeutics committee.
A public sector example was used to display how six sigma methodologies can be extended to cater for goals that are not primarily cost reduction. Reduction in defect and cost reduction are not mutual concepts in the short term. Customer has been substituted for patient, and reduced cost for successful therapy. Although not explicit, the case study did suggest that in the long term costs would be abridged through a reduced amount of administration time and investigation into faults/ compensation. Thus all Six Sigma projects have long term cost reduction consequences, however this is not always (but mostly) the motivation for implementation (as with the treatment of life threatening diseases).
The case study demonstrates the importance of the measurement of all major inputs into performance in order analyse how a process can be improved. Six Sigma stresses this measurement opposed to theoretical conjecture; it is “management by facts, not emotion” (Randall, R. cited in Erwin, 2001, p38)
Application to Services- Six Sigma Influence
Six Sigma is undeniably more complicated to apply in some service situation than those in manufacturing. Even where a process and goal exists some may argue that the setting of the specification limits can be somewhat a subjective issue and sometimes organisations spend time and money adding a specification value where one is not appropriate (Breyfogle et al, 2001, p196). This may be overcome by implementing a measurement systems analysis (MSA), however it must be noted that due to such issues, in services primary tasks may take longer than anticipated due to determining the appropriate measurement systems. (Breyfogle et al, 2001, p196).
This does not however mean that Six Sigma is not useful or is too difficult to implement- the extent of use and thus difficulty is dependent on company objectives. The methodology can be used to bring quick financial savings early on by tacking what Breyfogle coins the obvious ‘low hanging fruit’ problems in an organisation. By contrast it can also serve as a model for organisational culture “whereby everyone at all levels has a passion for continuous improvement with the ultimate aim of achieving virtual perfection” (cited from Basu & Wright, 2003, p3)
Some writers also maintain that various types of service industry are unsuitable for such rigid methodology as it will hinder the very facets that create customer satisfaction. There is often a trade off between customer satisfaction and running a business efficiently. For example a hairdressers may lose clients if it merely tried to fit as many haircuts in as possible (assuming no decline in haircutting quality), the customer in such circumstances like to be pampered, for the hairdresser to take their time and a relaxed atmosphere be upheld. Powell (1995) found that success derived more from HR intangibles, such as an open organisational culture, employee empowerment and executive commitment than on improved measurement, process improvement and benchmarking
This also links into the concept of reducing variability to decrease defect and increase efficiency. Although primarily founded on manufacturing quality, some services take this route- e.g. the mechanised “have a nice day” script in fast food chains etc. It is important to note that this will not lead to customer satisfaction in such sectors.
Process or Goal?
Behara et al (1995) state that in the early 1990s companies in the US (summary of all industries) were operating at around a three /four sigma quality level. The question is do companies need to reach Six Sigma level and is it in their best interests to do so? Initially one may believe that zero defects or total customer satisfaction is the ultimate goal that all companies should strive for (even if just for motivational purposes) as conveyed by the principles of Crosby. However understanding the traditional view of the trade-off between costs and prevention of service failures adds a different perspective. This concept is based on the premise that error prevention costs increase as the level of quality increases; in fact the relationship is exponential increases in prevention costs for mere incremental quality gains. Thus the target quality level managers should endeavour for may be under 100% and variable for different services dependent on their nature. (Heskett et al, 1990).
Six Sigma does not necessarily need to be achieved (Hammer & Goding, 2001), merely its methodology followed and an understanding of the optimal levels for overall cost reduction should be understood and set as the goal.
Among the literature, some authors have debated that Six Sigma is the latest fad, and that consists of a ‘repackaging’ of what has already come previously (Dusharme www.qualitydigest.com).
The challenge of Six Sigma is to overcome the ‘Innovative Fatigue’ (cited in Basu & Wright, 2003) which can cause loss of interest in an initiative. It has been shown by Turner (1993) that any quality initiative must be reinvented at regular intervals in order to sustain motivational levels of employees and that the maintenance and implementation of a quality program is approximately 2.5 years.
Improvement initiatives often forgo their initial success and do not gather the momentum necessary for true permanent organisational change for various hidden reasons. “Six Sigma is a quality approach that takes a whole system approach to improvement of quality and customer service so as to improve the bottom line” (Basu and Wright, 2003, p2) The main concept at this juncture is the ‘bottom line’ or return on assets as the key measure of success. This is a historical measurement that inherently can only inform of the result after it is too late to influence it. In many cases this may be too late and formal periodic assessments must be made in order to enable the flexibility to respond to various pressures. The ‘Control’ variable in the DMAIC methodology should ensure longevity and suppleness, and the DMADV (Define, Measure, Analyse, Design, Verify) methodology will serve to update and sustain processes also. Thus the question of whether Six Sigma is fashion or here to stay will only be answered through time.
Alternatives and improvements
It is also worth mentioning how Six Sigma has expanded and developed to illustrate its evolution is business and particularly services. Lean Six Sigma focuses on the maximisation of process velocity and provides tools for analysing the delay times and process flow for activities (George, 2003). It aims to reduce work in process and waste in procedures. Fit sigma (Basu & Wright, 2003) adds the element of sustainability and focuses not on the perfect goal of 3.4 defects per million but whatever the right ‘fit’ is for the organisation.
Finally Human Sigma does not focus so much on eliminating error, rather in reducing variance in key employee and customer outcomes, on the assumption that high variance equates to inefficient management. (Coffman, 2003). It seems that so many adaptations and variations of quality initiatives are being introduced due to the fact that organisations, particularly services are different in structure, ethics, goals etc. There does not seem to be one ‘best fit’ model and thus it is the predicament of the company to pick the one that suits it best.
As discussed in this essay there are many issues that must be considered when assessing whether to implement six sigma in services. These range from how one defines quality, identifies the costs of poor quality, implements statistical techniques to measure the situation, decides the level of sigma which will be optimal for the particular service industry they operate in etc. Despite these considerations one believes that Six Sigma is a useful tool in services, perhaps a reason why it has been criticised is that people have taken too literal an interpretation
It provides companies with a common metric that can be used across and organisation from production to customer satisfaction. It also presents one with the opportunities to compare results year on year, benchmark against rival firms and set goals for business evolution. Generally speaking, a higher sigma represents fewer errors and higher customer satisfaction (Behara et al, 1994).
The facts are that in the business world it is results that count and in this respect Six Sigma has been very successful (Hammer & Goding, 2001)
Asher, J.M., (1987), “Cost of Quality in Service Industries”, International Journal of Quality and Reliability Management, 5:5, pp38-46.
Bank, J., (2000), The Essence of Total Quality Management, FT/Prentice Hall, Harlow.
Basu, R. & Wright, J.N., (2003), Quality beyond Six Sigma, Butterworth-Heinemann, Oxford.
Behara, R.S., Gundersen, D.E., (2001), Analysis of Quality Management Practices in Services, International Journal of Quality and Reliability Management. 18:6, pp584-603.
Behara, R.S., Fontenot, G.F., Greysham, A., (1994), “Customer Satisfaction and analysis using six sigma”, International Journal of Quality and Reliability Management, 12:3, pp9-18.
Dale, B. G., (1999), Managing Quality, Blackwell Publishers, Oxford.
Deming, W.E., (1986), Out of the Crisis, Cambridge University Press, Cambridge.
Eckes, G., (2003) Six Sigma for Everyone, John Wiley & Sons, New Jersey.
George, M.L., (2003), Lean Six Sigma for Services, McGraw Hill, USA.
Ghobadian, A., Speller, S., Jones, M., “Service Quality- Concepts and Models”, International Journal of Quality and Reliability Management, 11:9, pp43-66.
Hammer, M., Goding, J., (2001), “Putting Six Sigma in Perspective”, Quality, 40:10, pp58.
Harry, M.J., Schroeder, R., (2000), Six Sigma: The Breakthrough Management Strategy Revolutionising the World’s Top Corporations, Doubleday, NY.
Heskett, J. l., Earlsasser, W., Hart, C.W.L., (1990), Service Breakthroughs: Changing the Rules of the Game, Macmillon Inc. USA.
Kooy, M.V., Edell, L, Melchiorre-Scheckner, H., (2002), “Use of Six Sigma to improve the Saftely and Efficacy of Acute Anticoagulation with Heparin”, Journal of Clinical Outcomes Management, 9:8, pp445-453.
Lagrasen, S., Lagrosen, Y., (2003) “Management of service quality- differences in values, practices and outcomes”, Managing Service Quality, 13:5, pp370-381.
Linderman, K., Schroeder, R. G., Zaheer, S., Choo, A.S., (2001), “Six Sigma: A Goal Theoretic Perspective”, Journal of Operations Management, 21:2, pp193-203.
McAdam, R., Canning, N. (2001), “ISO in the service sector: perceptions of small professional firms”, Managing Service Quality, 11:2, pp80-92.
Oakland, J.S. (1989), Total Quality Management, Heinemann Professional, Oxford.
Powell, T.C. (1995), “Total Quality Management as competitive advantage: a review and empirical study”, Strategic Management Journal, 16:1, pp15-37.
Reichheld, F., Sasser, W., (1990), “Zero Defections: Quality comes to services”. Harvard Business Review, Sept-Oct, pp105-11.
Robinson, L., (2003), “Committed to Quality: the use of quality schemes in UK public leisure services”, Managing Service Quality, 12:3, pp247-55.
Turner, J.R., (!999), The handbook of Project Based Management: Improving the process for achieving strategic objectives. 2nd Ed. McGraw Hill, London.
White, F.M., Locke, E. A., (1981), “Perceived Determinants of high and low productivity in three occupational groups: a critical incident study” Journal of Management Studies, 18, pp375-387.
Wood, M., (1994), “Statistical Methods for Monitoring Service Processes”, International Journal of Service Industry Management, 5:4, pp53-69.
Zeithaml, V.A., (1981),. “How Consumer evaluation processes differ between goods and Services” in Donnelly, J. and George, W. (Eds) Marketing of Services. American Marketing Association, Chicago, pp186-190.
Coffman, C. (2003), HumanSigma, Managing the human Difference. Gallup Management Journal.
Dusharme, D., “Survey: Six Sigma Packs a Punch”.
Erwin, J. (2001), “Flawless”, Quality World, Jan ed.
Erwin, J. (2000), “Achieving Total Customer Satisfaction through Six Sigma”, Quality Digest.
Smith, K., “Six Sigma for the service sector”