Six Sigma – so what?

Citation

Morgan, J. (2006), "Six Sigma – so what?", Assembly Automation, Vol. 26 No. 1. https://doi.org/10.1108/aa.2006.03326aaa.002

Six Sigma – so what?

Six Sigma – so what?

Keywords: Quality control

Six Sigma can be viewed as a number of different things, but from your customers perspective it means you are getting it right. And you are getting it right virtually every time, with less than 4 defects in every million activities. Six-Sigma equates to 3.4 defects per million opportunities (DPMO). And the “defects” are aspects of your process, product, or service that the customer has identified as important.

It should also be remembered that in a manufacturing environment that “the customer” includes all the assembly stages further down the line. So in a sense a company is its own customer and it is only the final product that has a third party as a “customer”.

Imagine the improvement in the effectiveness and efficiency of your organisation if you were to achieve Six Sigma performance, as viewed by your “customers”.

For me, Six Sigma is a rigorous quantitative approach to quality improvement. It is an initiative that pulls together a number of statistical and creative tools and techniques to produce an all embracing approach to the improvement of processes, products and services. And it works!

There are a number of key elements in achieving Six Sigma:

• •

  firstly, we must genuinely focus on the customer and identify their critical to quality requirements, known as CTQs;

• •

  we must ensure that our processes are designed and managed to meet these CTQs;

• •

  we must ensure that we have appropriate measurement in place to understand how well we are meeting the customer requirements and how well the customer feels we are meeting the requirements;

• •

  we need to involve our people, ensuring they are effectively equipped so that they are able and feel able to challenge their processes and improve the way they work;

• •

  and we need to undertake that improvement using a systematic problem solving and process improvement approach.

Naturally, none of this happens without the right leadership and management behaviour. This quote from Albert Einstein says much “The significant problems we face cannot be solved by the same level of thinking which caused them.” Six Sigma thinking is what is needed; and former GE CEO Jack Welch demonstrates that thinking in so many ways. As he says “We are going to shift the paradigm from fixing products to fixing and developing processes, so that they produce nothing but perfection or close to it!!” Let us start to put the approach into perspective with an example. Through customer research you have identified that your customers have key requirements from your organisation. Typically, these will relate to speed, accuracy, or completeness in some way. In assembly, there the requirements might also include things like size and tolerances, or perhaps the position and location of parts. In this example, we will keep things simple and assume that your customers have only one CTQ – your product is issued within five working days of their order.

Meeting this key requirement means you have a defect free transaction, but if you take longer than five days you will have failed and produced a defect. So, now lets imagine you have measured the time it takes to issue the products and have presented the data as a histogram, as shown in Figure 1.

Figure 1 Histogram showing the time it takes to issue products to customers

The pattern of the results shows a normal distribution, and we can see a “bell-shaped curve”. With this data we can work out the standard deviation. The standard deviation gives us the average distance from the mean for any one of the products issued. And for standard deviation, read process s when viewed against the customers' CTQs. Figure 2 shows the different defect levels for the various s values from one through to six.

Figure 2 We can take the issue time data and calculate the standard deviation 5 = σ. The diagram shows the likely percentage of cases within each band of σ values

Six Sigma performance means you are meeting the customer requirements 99.999998 per cent of the time. We can convert these values into DPMO. Figure 3 shows the s and DPMO values, but you will note that these have been statistically adjusted.

Figure 3 We translate the s values into DPMO, but in doing so, we make an adjustment to account for the effects of on-going variation

The adjustment looks to make allowance for the effects of on-going variation in the results, caused by changes in the various elements involved in the process. These will include people, equipment, methods, materials and environment. Without this adjustment, Six Sigma would equate to 0.002 DPMO.

Returning to our example, imagine that we now introduce the distribution curve identifying the s values from one through to six. By comparing these to the customers' key to quality requirement of five days, we can identify that we have a large number of cases in the tail outside the customer specification. These are all defects as Figure 4 shows.

Figure 4 We can see the sigma performance of our process by looking at the customer specification. The defects “in the tail” highlight that our process is not capable. We need to reduce the variation

Our objective now must be to improve the s performance of our process by “squeezing in” the bell shaped distribution curve as shown in Figure 5. In other words, by reducing variation.

Figure 5 In this case we need to move our performance by shifting the mean and reducing variation. And we need to keep doing it so our distribution curve gets tighter and tighter

There is a range of systematic approaches for problem solving and process improvement, be they four, five, six, seven or eight steps or even more. The most common is define, measure, analyse, improve, control (DMAIC). The key is that the approach is genuinely systematic and relies on data to ensure correct decisions are made. All too often, managers jump to conclusions about what a problem is and how to fix it. And all too often, they are wrong!

Measurement is at the heart of the improvement journey to Six Sigma. It may have prompted the need for a process improvement, perhaps because we have identified that s performance is low or because of customer complaints. And we will see measurement coming into play throughout the improvement cycle, helping us to understand the current situation and verify possible causes, for example.

The improvement projects undertaken in a Six Sigma initiative will clearly link to the organisation's strategic ambitions and business plans. The efforts are firmly focused on improving customer satisfaction and the bottom line.

So many TQM initiatives failed to do this and withered on the vine as a result.

John MorganBased at Catalyst Consulting Limited