Management Tools: TQM Related Topics Continuous Improvement Malcolm Baldrige National Quality Award Quality Assurance Six

Six Sigma Description Total Quality Management (TQM) is a systematic approach to quality improvement that marries product and service specifications to customer performance. TQM then aims to produce these specifications with zero defects. This creates a virtuous cycle of continuous improvement that boosts production, customer satisfaction and profits. Methodology In order to succeed, TQM programs require managers to: Assess customer requirements Understand present and future customer needs; Design products and services that cost-effectively meet or exceed those needs. Deliver quality Identify the key problem areas in the process and work on them until they approach zero-defect levels; Train employees to use the new processes; Develop effective measures of product and service quality; Create incentives linked to quality goals; Promote a zero-defect philosophy across all activities; Encourage management to lead by example; Develop feedback mechanisms to ensure continuous improvement. Common Uses TQM improves profitability by focusing on quality improvement and addressing associated challenges within an organization. TQM can be used to: Increase productivity; Lower scrap and rework costs; Improve product reliability; Decrease time-to-market cycles; Decrease customer service problems; Increase competitive advantage.

Selected References Besterfield, Dale H., Carol Besterfield-Michna, Glen Besterfield, and Mary BesterfieldSacre. Total Quality Management, 3d ed. Prentice Hall, 2002. Camison, Cesar. "Total Quality Management and Cultural Change: A Model of Organizational Development." International Journal of Technology Management, Vol. 16, No. 4/5/6, 1998, pp. 479-493. Choi, Thomas Y., and Orlando C. Behling. "Top Managers and TQM Success: One More Look After All These Years." Academy of Management Executive, February 1997, pp. 37-47. Dahlgaard, Jens J., Kai Kristensen, and Ghopal K. Khanji. Fundamentals of Total Quality Management. Routledge, 2005. Deming, W. Edwards. Quality, Productivity, and Competitive Position. MIT Press, 1982. Feigenbaum, Armand V. Total Quality Control, 4th ed. McGraw-Hill, 1991. Gale, Bradley T. Managing Customer Value: Creating Quality and Service That Customers Can See. Free Press, 1994. Goetsch, David L., and Stanley B. Davis. Quality Management: Introduction to Total Quality Management for Production, Processing, and Services, 5th ed. Prentice Hall, 2005. Grant, Robert M., Rami Shani, and R. Krishnan. "TQM's Challenge to Management Theory and Practice." Sloan Management Review, Winter 1994, pp. 25-35. Imai, Masaaki. Kaizen: The Key to Japan's Competitive Success. McGraw-Hill, 1989. Juran, J.M. Juran on Quality by Design: The Next Steps for Planning Quality into Goods and Services. Free Press, 1992. Malcolm Baldrige National Quality Award, 2006 Award Criteria. http://www.quality.nist.gov. Walton, Mary. The Deming Management Method. Perigree, 1986. Lean Six Sigma Related topics Lean Manufacturing Six Sigma

Statistical Process Control Total Quality Management

Description Lean Six Sigma combines elements of both Lean Manufacturing and Six Sigma approaches. Lean Six Sigma originally was designed to improve manufacturing quality to no more than 3.4 defects per million opportunities. The combination of Lean and Six Sigma helps companies achieve higher quality in a fast and efficient way by creating a culture of responsiveness and accountability. Lean Six Sigma programs constantly measure and analyze data on the variables in any process, then use statistical techniques to understand what improvements will reduce defects and improve efficiency. Such programs also incorporate a strong system for gathering customer feedback. Companies have applied Lean Six Sigma to functions ranging from manufacturing to call centers to collections. Methodology Prior to using Lean Six Sigma, companies should deploy an upfront diagnostic to identify the most critical opportunities. Then, Lean Six Sigma teams follow five problem-solving steps to quickly identify root problem causes, develop solutions and put in place procedures that maintain those solutions. Define. Identify the customer requirements, clarify the problem and set goals; Measure. Select what needs to be measured, identify information sources and gather data; Analyze. Develop hypotheses and identify the key variables and root causes; Improve. Generate solutions and put them into action, either by modifying existing processes or by developing new ones. Quantify costs and benefits; Control. Develop monitoring processes for continued high-quality performance. Common uses Companies use Lean Six Sigma to set performance goals for the entire organization and to mobilize teams and individuals to achieve dramatic improvements in existing processes. More specifically, Lean Six Sigma can: Make processes more rigorous and efficient by using hard, timely data to make operating decisions; Cultivate customer loyalty by delivering superior value; Accustom managers to operating in a fast-moving internal business environment that mirrors marketplace conditions; Achieve quantum leaps in product performance; Reduce variation in service processes, like time from order to delivery; Improve financial performance through cost savings, increased revenue and expanded operating margins.

Selected references Breyfogle, Forrest, III. Implementing Six Sigma: Smarter Solutions Using Statistical Methods, 2d ed. John Wiley & Sons, 2003. Devane, Tom. Integrating Lean Six Sigma and High-Performance Organizations: Leading the Charge Toward Dramatic, Rapid, and Sustainable Improvement. Pfeiffer, 2003. Eckes, George. The Six Sigma Revolution: How General Electric and Others Turned Process into Profits. John Wiley & Sons, 2001. George, Michael L. Lean Six Sigma for Service: How to Use Lean Speed and Six Sigma Quality to Improve Services and Transactions. McGraw-Hill, 2003. Hariharan, Arun. "CEO's Guide to Six Sigma Success." ASQ Six Sigma Forum Magazine, May 2006, pp. 16-25. Preis, Kim H. Six Sigma for the Next Millennium: A CSSBB Guidebook. American Society for Quality, 2005. Snee, Ronald D., and Roger W. Hoerl. Leading Six Sigma: A Step-by-Step Guide Based on Experience with GE and Other Six Sigma Companies. Financial Times Prentice Hall, 2002. Sodhi, ManMohan S., and Navdeep S. Sodhi. "Six Sigma Pricing." Harvard Business Review, May 2005, pp. 135-142. Taghizadegan, Salman. Essentials of Lean Six Sigma. Butterworth-Heinemann, 2006. Wedgwood, Ian D. Lean Sigma: A Practitioner's Guide. Prentice-Hall PTR, 2006

What is Six Sigma?

The term sigma is a Greek alphabet letter () used to describe variability. In Six Sigma, the common measurement index is DPMO (Defects Per Million Operations) and can include anything from a component, piece of material, or line of code, to an administrative form, time frame or distance. A sigma quality level offers an indicator of how often defects are likely to occur, where a higher sigma quality level indicates a process that is less likely to create defects. Consequently, as sigma level of quality increases, product reliability improves, the need for testing and inspection diminishes, work in progress declines, cycle time goes down, costs go down, and customer satisfaction goes up.

To have a more comprehensive understanding about sigma quality level, it will be explained from two perspectives of process capability: short-term and long-term process capabilities. Short-term process capability A part or item is classified as defective if the desired measurement, denoted by X, is outside the customer-supplier specification limit (USL) or lower specification limit (LSL). In addition to specifying the USL and LSL, a customer would also specify a target value, which typically is the midpoint between the USL and LSL. From a short-term process capability view, after sampling data from the process, a six sigma process that produces the parts is normally distributed (see Figure 2.1). Table 2.1 displays short-term process capability in various sigma levels. Table 2.1: Short-Term Process Capability at Various Sigma Quality Levels

Sigma Level % Good 2 3 4 5 6 95.45 99.73 99.9937 99.999943

PPM/DPMO 45500 2700 63 0.57

99.9999998 0.002

Figure 2.1: Short-Term Six Sigma Performance for a Single Process

Long-term process capability Due to the nature of the process, when dealing with the situation of a long-term process, shifts and drifts in the mean of the distribution of a component value occur for a number of reasons as do changes in other parameters of the distribution: for example, tool wear is one source of a gradual drift, differences in raw material or change of suppliers can cause shifts in the distribution. A solution proposed by D.H. Evans (Statistical Tolerancing: The State of the Art Part III, Shifts and Drifts 1975) focuses on high production rates, and low cost components. Evans suggests that one should use 1.5s as the standard deviation to calculate the percentage of out of tolerance responses. Furthermore, research by M.J. Harry (The Nature of Six Sigma Quality 1988) has shown that a typical process is likely to deviate from its natural centering condition by approximately 1.5 standard deviations at any given moment in time. With this principle in hand, one can make a rational estimate of the long-term process capability with knowledge of only the short-term process capability (see Figure 2.2). Table 2.2 displays long-term process capability in various sigma levels. Figure 2.2: Long-Term Six Sigma Performance for a Single Process (Shifted 1.5)

Table 2.2: Long-Term Process Capability in Various sigma Levels

Sigma Level % Good 2 3 4 5 6 69.15 93.32 99.379 99.9676

PPM/DPMO 308,537 66,807 6,210 233

99.99966 3.4

INTRODUCTION TO SIX SIGMA

Six Sigma is usually related to the magic number of 3.4 defects per million opportunities. People often view Six Sigma as yet another rigorous statistical quality control mechanism.

Pioneered at Motorola in the mid-1980s, Six Sigma was initially targeted to quantify the defects occurred during manufacturing processes, and to reduce those defects to a very small level. Motorola claimed to have saved several million dollars. Another very popular success was at GE. Six Sigma contributed over US $ 300 million to GE's 1997 operating income. Today Six Sigma is delivering business excellence, higher customer satisfaction, and superior profits by dramatically improving every process in an enterprise from financial to operational to production. Six Sigma has become a darling of a wide spectrum of industries, from health care to insurance to telecommunications to software. What is Six Sigma? It is important to recall that every customer always values consistent and predicable services and/or products with near zero defects. Therefore they experience the variation and not the mean. Mean is their expectation or our target. If we can measure process variations that cause defects i.e. unacceptable deviation from the mean or target, we can work towards systematically managing the variation to eliminate defects. Six Sigma is a methodology focused on creating breakthrough improvements by managing variation and reducing defects in processes across the enterprise. Sigma is a Greek symbol represented by "". Why do we call Six Sigma as Six Sigma and not Four or Five Sigma or Eight Alpha (another Greek symbol)? Sigma is a statistical term that measures process deviation from process mean or target. Mean is also referred as average in common language. The figure of six was arrived statistically by looking at the current average maturity of most business enterprises. We would like to revise this figure to 8 or may be 9 provided the world becomes a more orderly and predictable (even with increasing entropy or chaos) place to live! We have a detailed discussion on keywords "breakthrough improvement" and "variation" apart from the "methodology" in later sections.

Example Let us take an example to bring a breakthrough improvement in our current understanding of the concept of Six Sigma. This requires us to have basic knowledge of statistics. We have a detailed discussion on required statistical concepts later. Consider a pizza delivery shop that guarantees the order delivery with 30 minutes from the time of accepting an order. In the event of a delivery time miss, the customer is refunded 100% money. The management took a target (read mean) of delivering every pizza order within 15 minutes and aligned its processes to meet this goal. If we collect data of delivery times over a large number of the delivery made by the pizza shop and make a frequency distribution graph, we discover that it resembles a "bell shaped curve". A frequency distribution graph is constructed from the frequency table; a frequency table lists different time intervals (called classes) like 0 to 2 minute, 2 to 4 minutes, to 14 to 16 minutes to 28 to 30 minutes and the count of the deliveries made in each interval. The mean is found to be 16 minutes and standard deviation (measure of deviation or dispersion in data i.e. ) is found as 2.5 minutes. A graph drawn from the data of over 5000 deliveries made is given below. Note, this not a real graph and is used only for illustration purposes.

This bell shape curved is called "normal distribution" in statistical terms. In real life, a lot of frequency distributions follow normal distribution, as is the case in the pizza delivery times. Natural variations cause such a distribution or deviation. One of the characteristics of this distribution is that 68% of area (i.e. the data points) falls within the area of -1 and +1 on either side of the mean. Similarly, 2 on either side will cover approximately 9

95.5% area. 3 on either side from mean covers almost 99.7% area. A more peaked curve (e.g. more and more deliveries were made on target) indicates lower variation or more mature and capable process. Whereas a flatter bell curve indicates higher variation or less mature or capable process. After this statistical detour let us come back to our pizza example. If the pizza shop delivers 68% of pizza orders in time, we call it a "One Sigma shop". Similarly, if the pizza shop makes 95.5% deliveries on time, we call it a "Two Sigma shop". In our example, data suggests that it is almost a "Three Sigma shop". We should now be able to appreciate why management took a delivery time target of 15 minutes and not 30 minutes. Imagine what would have happened with a 30 minutes delivery time target! The "delivery time" is a critical-to-quality parameter from the customer perspective and has a significant impact on profits. In addition, it is an entry barrier for the competition. Such a parameter is called a CTQ and its definition in context of our pizza shop is given below: CTQ Name: Timely Pizza delivery CTQ Measure: Time in Minutes CTQ Specification: Delivery with 30 minutes from the order acceptance time Now we can easily define a defect: Defect: Delivery that takes longer than 30 minutes Unit: Order Opportunity: 1 per order i.e. only "1" defect can occur in "1" order Technical Note: This discussion on the example did not include 1.5 process shift during the above analysis. The concept is discussed later. The Six Sigma conversion graph including a 1.5 shift in process is given below:

10

This graph is on a logarithmic scale. Notice the increasing rate of improvement. For example, 1 sigma to 3 sigma is only 10 times improvement; 3 sigma to 4 sigma is a big 10 times improvement; whereas 5 sigma to 6 sigma is a whooping 1825 times change. That is why we are talking about breakthrough improvements in a journey to Six Sigma. How does Six Sigma work? The driving force behind any Six Sigma project comes from its primary focus - "bringing breakthrough improvements in a systematic manner by managing variation and reducing defects". This requires us to ask tougher questions, raise the bar significantly, and force people to think out of the box and to be innovative. The objective is to stretch, stretch mentally and not physically. To make this journey successful there is a methodology(s) to support Six Sigma implementations. There are 2 potential scenarios - (a) there is already an existing process(s) that is working "reasonably" well and (b) there is no process at all. A bad process is as good as no process. Scenario (a) focuses on significant process improvements and requires use of DMAIC: 1. Define process goals in terms of key critical parameters (i.e. critical to quality or critical to production) on the basis of customer requirements or Voice Of Customer (VOC) 2. Measure the current process performance in context of goals 3. Analyze the current scenario in terms of causes of variations and defects 4. Improve the process by systematically reducing variation and eliminating defects 5. Control future performance of the process

11

Scenario (b) focuses on process design using Design For Six Sigma (DFSS) approach. DFSS typically requires IDOV: 1. Identify process goals in terms of critical parameters, industry & competitor benchmarks, VOC 2. Design involves enumeration of potential solutions and selection of the best 3. Optimize performance by using advanced statistical modeling and simulation techniques and design refinements 4. Validate that design works in accordance to the process goals Note, sometimes a DMAIC project may turn into a DFSS project because the process in question requires complete re-design to bring about the desired degree of improvement. Such a discovery usually occurs during improvement phase of DMAIC.

An Introduction to Six Sigma Management

Author: Howard Gitlow Posted: 10/30/2008 2:55:44 PM EDT Non-Technical Definitions of Six Sigma Management Six Sigma management uses statistical process control to relentlessly and rigorously pursue the reduction of variance and standard deviation in all critical processes to achieve continuous and breakthrough improvements that impact the bottom-line and/or top-line of the organization and increase customer satisfaction. Another common definition of Six Sigma management is that it is an organizational initiative designed to create manufacturing, service and administrative processes that produce a high rate of sustained improvement in both defect reduction and cycle time (e.g., when Motorola began their effort the rate they chose was a 10-fold reduction in defects in two years, along with a 50 percent reduction in cycle). For example, a bank takes 60 days on average to process a loan with a 10 percent rework rate in 2000. In a Six Sigma organization, the bank should take no longer than 30 days, on average, to process a loan with a 1 percent error rate in 2002, and no more than 15 days, on average, to process a loan with a 0.10 percent error rate by 2004. Clearly, this requires a dramatically improved/innovated loan process. Technical Definition of Six Sigma Management The Normal DistributionThe term Six Sigma is derived from the normal distribution used in statistics. Many observable phenomena can be graphically represented as a bellshaped curve or a normal distribution as illustrated in Figure 1.

12

Figure 1: Normal Distribution with Mean (m=0) and Standard Deviation (s=1) When measuring any process, it can be shown that its outputs (services or products) vary in size, shape, look, feel or any other measurable characteristic. The typical value of the output of a process is measured by a statistic called the mean or average. The variability of the output of a process is measured by a statistic called the standard deviation. In a normal distribution, the interval created by the mean plus or minus two standard deviations contains 95.44 percent of the data points, or 45,600 data points per million (or sometime called defects per million opportunities denoted DPMO) are outside of the area created by the mean plus or minus two standard deviations [(1.00-.9544= . 0456)x1,000,000=45,600]. In a normal distribution the interval created by the mean plus or minus three standard deviations contains 99.73 percent of the data, or 2,700 DPMO are outside of the area created by the mean plus or minus three standard deviations [(1.00-.9973=.0027)x1,000,000 = 2,700]. In a normal distribution the interval created by the mean plus or minus six standard deviations contains 99.9999998 percent of the data, or two data points per billion data points outside of the area created by the mean plus or minus six standard deviations. Six Sigma management promotes the idea that the distribution of output for a stable normally distributed process (Voice of the Process) should be designed to take up no more than half of the tolerance allowed by the specification limits (Voice of the

13

Customer). Although processes may be designed to be at their best, it is assumed that over time the processes may increase in variation. This increase in variation may be due to small variation with process inputs, the way the process is monitored, changing conditions, etc. The increase in process variation is often assumed for the sake of descriptive simplicity to be similar to temporary shifts in the underlying process mean. The increase in process variation has been shown in practice to be equivalent to an average shift of 1.5 standard deviations in the mean of the originally designed and monitored process If a process is originally designed to be twice as good as a customer demands (i.e., the specifications representing the customer requirements are six standard deviations from the process target), then even with a shift, the customer demands are likely to be met. In fact, even if the process shifted off target by 1.5 standard deviations there are 4.5 standard deviations between the process mean (m + 1.5s) and closest specification (m + 6.0s), which result in at worst 3.4 DPMO at the time the process has shifted or the variation has increased to have similar impact as a 1.5 standard deviation shift. In the 1980s, Motorola demonstrated that a 1.5 standard deviation shift was in practice was observed as the equivalent increase in process variation for many processes that were benchmarked. Figure 2 shows the Voice of the Process for an accounting function with an average of seven days, a standard deviation of one day, and a stable normal distribution. It also shows a nominal value of seven days, a lower specification limit of four days, and an upper specification limit of 10 days. The accounting function is referred to as a three-sigma process because the process mean plus or minus three standard deviations is equal to the specification limits, in other terms, USL= +3 and LSL = 3. This scenario will yield 2,700 defects per million opportunities or one early or late monthly report in 30.86 years [(1/0.0027)/12].

14

Figure 2: Three Sigma Process with 0.0 Shift in the Mean

Figure 3 shows the same scenario as in Figure 2, but the process mean shifts by 1.5 standard deviations (the process average is shifted down or up by 1.5 standard deviations [or 1.5 days] from 7.0 days to 5.5 days or 8.5 days) over time. This is not an uncommon phenomenon. The 1.5 standard deviation shift in the mean results in 66,807 defects per million opportunities, or one early or late monthly report in 1.25 years [(1/.066807)/12].

15

Figure 3: Three Sigma Process with a 1.5 Sigma Shift in the Mean Figure 4 shows the same scenario as Figure 2 except the Voice of the Process only takes up half the distance between the specification limits. The process mean remains the same as in Figure 2, but the process standard deviation has been reduced to one half-day through application of process improvement. In this case, the resulting output will exhibit 2 defects per billion opportunities, or one early or late monthly report in 41,666,667 years [(1/.000000002)/12].

16

Figure 4: Six Sigma Process with a 0.0 Shift in the Mean Figure 5 shows the same scenario as Figure 4, but the process average shifts by 1.5 standard deviations (the process average is shifted down or up by 1.5 standard deviations [or 0.75 days = 1.5x0.5 days] from 7.0 days to 6.25 days or 7.75 days) over time. The 1.5 standard deviation shift in the mean results in 3.4 defects per million opportunities, or one early or late monthly report in 24,510 years [(1/.0000034/12]. This is the definition of Six Sigma level of quality.

17

Figure 5: Six Sigma Process with 1.5 Sigma Shift in the Mean References: HowardGitlow.com Gitlow, H., Oppenheim, A., Oppenheim, R. and Levine, D., Quality Management: Tools and Methods for Improvement, 3rd ed. (New York: McGraw-Hill-Irwin, 2004)

18