logo_banner

Scottish Technical Meeting

Friday, 22 February 2013

Branch President George Beattie FICME, pictured right, opened the first meeting of 2013 by wishing everyone a Happy New Year, and welcomed the speaker, Dr Rajesh Ransing FICME from Swansea University, who then presented his paper “ Beyond six Sigma, Zero Defect Casting Manufacture “

Dr Ransing explained that 7 Epsilon was developed to address some of the shortcomings of 5 S, 6 sigma and 8D processes and that its promotion was being led by Swansea University. He went on to say that one of the main aims is to encourage knowledge retention and it’s reuse to help improve manufacturing processes. It was stated that currently little of the masses of data collected by companies was in a format that could be readily used for the processes listed above. He suggested that thought be given to future methods of data collection so that it is more readily usable.

He went on to explain that falls in scrap rates from relatively high levels to lower levels tend to be attributed to a few factors easily identifiable, but to achieve reductions at lower levels, which is not so easy, generally a number of factors may have an influence all interacting with each other. This is when it becomes difficult to assess which parameters have the greatest influence, by using 7 Epsilon this can help identify the important parameters.
Given a list of measurable factors and their ranges, 7 Epsilon can be applied, whereby influencing factors can be identified from the analysis of the data using penalty matrices to generate bubble diagrams, which identify the main effects and interactions. Process control improvements can then be developed from this information leading to a reduction in process variation, this information being used for continuous improvement.

Dr Ransing demonstrated 7 Epsilon, using a significant amount of the data collated for a nickel alloy casting. He took us through the various steps which culminated in identifying the influencing factors which helped with scrap reduction.

Because of the need to have significant amounts of data to analyse, it could be thought that the 7 Epsilon approach would be suitable only 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 using 7 Epsilon can be made.

There followed a lively question and answer session before the Branch Vice President Bobby Brown FICME gave the vote of thanks.