7 Epsilon, Beyond 6 Sigma Thinking

Objective Zero Defect Casting Manufacture

Friday, 7 September 2012

Branch President Steve Smith FICME welcomed both members and visitors present to this the first technical lecture of the new presidential year before introducing Dr. Rajesh Ransing FICME of Swansea University who was to speak about a new method of tackling manufacturing process improvements by the 7 Epsilon approach.

Since the presentation Dr Ransing has now posted some videos on you-tube that members can watch, the link is below at the foot of this article and can be copied and pasted into your browser.

Dr. Ransing explained that this talk was to introduce the philosophy of 7 Epsilon and was part of a road show which has already been presented in several countries. A more detailed presentation would be available on YouTube within 2-3 weeks from the date of this talk. An internet based 'on demand' course to teach the philosophy will be offered by both CMI and ICME later in the year.

7 Epsilon was developed to address some of the shortcomings of 5s, 6 Sigma and 8D, and now the push to bring the philosophy to manufacturers is being led by Swansea University. The aim is to engender knowledge retention and its re-use to gain new knowledge of manufacturing processes and develop a culture of innovation within the manufacturing unit.

Any particular product may be produced at more than one manufacturing centre. The process to produce it satisfactorily will vary at each centre. It is not possible to say that the particular process parameters at any one centre can be successfully transferred to another as each centre is unique and therefore improvements in the process must be specific to that centre. The 7 Epsilon philosophy is therefore to develop greater understanding of your process and optimise it to achieve your goal whether that be defect reduction, enhancing mechanical properties of cast components, improve margins or for any other technical benefit.

The ethos of 7 Epsilon is to obtain actionable process information in the form of a list of measurable factors and their ranges in order to generate improvements to meet a desired business goal which may be either financial or technical. Adoption of the philosophy will, by its very nature of reviewing process data, systematically analysing it, making improvements and involving everyone who could influence the process, bring the philosophy into the culture of the company. Companies that have decided to follow this route have very quickly seen benefits; however some have fallen at the first hurdle, and not gone on to adopt 7 Epsilon, because of having to pull together all the relevant process data because this is a necessary but often time consuming essential first step.

The current approach is to decide on a target, e.g. 10% reduction in scrap rate. Then design experiments possibly using computer simulation to review the potential result with own foundry personnel and/or industry experts. Very often it can become a series of trial and error as the data available is generally fragmented and is rarely reused. Access to company/product specific process knowledge and previous published literature in foundry conferences/journals is not available at the point of need. As a result, determining the influencing factors is still an art requiring tens of years of experience.

The 7 Epsilon approach is to provide a mechanism whereby the influencing factors are realised from the analysis of the data using scatter diagrams then applying penalty matrices to generate bubble diagrams . This identifies the main affects and interactions among measurable factor ranges. Process control improvements can be developed from the resulting information which will lead to a reduction in process variations and more importantly generate new process knowledge which can then be reused to further reduce process variations for new or similar cast components. This philosophy of new knowledge gain and its reuse to continuously improve will help to take 7 Epsilon into the culture of the company.

At this point Dr Ransing provided a worked example to demonstrate the power of 7 Epsilon. The example was to find from a significant amount of data which were the influencing factors if scrap reduction is to be achieved in a nickel alloy casting.

Because of the need to have significant amounts of data to analyse it could be thought that the 7 Epsilon approach would only be suitable for repetition foundries but this is not the case. Jobbing foundries usually have jobs which have sufficient similarities to enable their data to be collated and by doing so generate the information from which improvements through 7 Epsilon can be made.

There followed a lively question and answer session before Professional Member John Small gave a vote of thanks to Dr. Ransing for a most interesting and thought provoking presentation.