Latest Website Additions (Six Sigma Training - Is it worth the investment? - By John Ryan)

Six Sigma Training – Is it worth the investment?
By John Ryan, Six Sigma Tutor, SQT Training Ltd.

Every year companies are faced with pressures to reduce costs, reduce defects and in turn, reduce budgets. So why in this tough economic climate, would a company willingly spend €12,500 on a training programme and how, as many Third Party Auditors ask, “can you measure the effectiveness of that training?”

In this article we look at how two companies took the decision to pursue Six Sigma Black Belt Training with SQT Training Ltd and how their initial expense, turned into a wise investment. The Black Belt programme involves 20 days training carried out over a 4 month period (1 week per month) with the delivery of a strategic company project, the key outcome.

Our first company, an electronics manufacturer in Ireland, sent two delegates on the programme, each armed with a key problem to be investigated. The first problem or "project" concentrated on increasing the reliability of a non-core but nonetheless necessary, screen manufacturing process. At the start of the first project the screen reject rate from the screen manufacturing process was in excess of 50% ; high quality screens could be purchased from a trusted Japanese source but at a high cost and with a 9 to 12 week lead time. The strategy taken was to use the Japanese screens as a Benchmark and compare their characteristics with those from the local process. To be comparable, the variation in the local screens would have to be reduced from a standard deviation of 5.8 to 3.1 (roughly 46%). By forming a team of process experts and utilising a number of analytical tools, the input variables were narrowed down from an initial total of 132, to a short list of just 11. A Design of Experiment (DOE) was completed using the four most likely variables at various settings or levels. From this experiment a linear model was derived which showed the relative impact of each of the variables on the standard deviation.

Using the new model the trainee Black Belt was able to determine the optimum process settings, which confirmed that the process standard deviation could be reduced by 40% resulting in a 20% unit cost reduction and an 80% lead time reduction. The resultant savings were in excess of €125,000.

Fig 1. Printed patterns from the screen manufacturing process show the before and after surface area variation.


Before: Wavy surface area (SD = 5.8)	After: 40% improvement (SD = 3.5)

The second project successfully reduced the defect level of specific components being shipped to one of their European customers from 611 ppm to approximately 45 ppm. This project proved more difficult as most of the defects were due to ‘special cause’ variation but using tools such as process mapping, failure mode and effects analysis and mistake proofing, ppm reduction was achieved and more importantly sustained, generating annual savings in excess of €114,000.

Our second company is Molex, based in southwest Ireland. Molex is the world's second-largest manufacturer of electronic, electrical and fiber optic interconnection products and systems. Molex also manufacture a variety of switches and application tooling. Molex has product development centers, manufacturing facilities and sales offices on six continents. The Irish plant were challenged by a 28,000 ppm internal reject level on a new product line. Initial investigations identified that resistance failures made up 70% of the problem. The second discovery was that the 28,000 ppm was only achieved by retesting ‘failed’ units (44,000 ppm from the first test). Therefore the focus was immediately drawn to the resistance test equipment. Using team tools such as process mapping, cause and effect matrices and process failure mode and effects analysis (FMEA), the source of many of the false rejects were identified. Two immediate changes were implemented bringing the ppm level down to 11,500 ppm. However, a measurement system study continued to puzzle the team as they struggled to get the three resistance testers to agree exactly in repeated tests. Following some research into the linearity of resistance measurement devices at low resistance levels (below 1 ohm), a breakthrough discovery was made. While resistance measurement was indeed linear above 1 ohm, when measurements fell below 1 ohm there was a non-linear relationship. The problem was solved by developing resistance standards of 0.33 ohms and 0.5 ohms to allow for more accurate calibration at lower resistance levels. By eliminating this linearity problem and a bias problem between the three testers, the team got all three to accurately repeat in a guage R&R study. The result was a drop to 4000 ppm, with all three testers giving comparable results. The trainee Black Belt’s diligence and perseverance resulted in an 85% reduction in reject level and savings of over €75,000 p.a.

Fig 2. Project performance showing dramatic fall in PPM

Chart : Project performance showing dramatic fall in PPM

Collectively these 3 projects delivered cost savings in excess of €300,000. So what about return on investment? Well, how many training programmes do you know can deliver a hundred-fold return on initial investment within the year? If you keep in mind that trained Black Belts should be capable of completing at least two major projects a year, then the return on investment should be more like 300-500%, even with an 80% project success rate.

Now that’s what I call a real measure of training effectiveness!

John Ryan, Six Sigma Tutor, SQT Training Ltd. SQT Training Ltd. has been providing training solutions to large multinationals and small indigenous companies since 1989. Our courses include Six Sigma, Statistical Problem Solving, Pharmaceutical & Medical Device Standards, Food Safety Management, Health & Safety, Quality, Environment and Leadership Development programmes - all available on both a Public and In-Company basis. For further details contact: Web: www.sqt.ie Tel: 061-339040 Email: info@sqt.ie

top of page