An AHP parsimonious based approach to handle manufacturing errors in production processes

This paper introduces a new hybrid tool to classify manufacturing errors in production processes. Our tool is a hybrid procedure based on the combination of the parsimonious AHP method with a clustering method. Parsimonious AHP permits to use a much smaller number of pairwise comparisons with respect to classical-AHP. At the same time, the use of a clustering method, allows for the assignment of the alternatives on the basis of their closeness to each other and not on the basis of some reference profiles as happens in several other methods. Furthermore, some veto systems help the decision maker to better define how the errors belong to different classes in a participative way. The approach has been tested in a largest automotive plants in Italy. This approach has provided the company with a participative and robust theoretically funded tool that supports the understanding of the seriousness of the manufacturing errors. From a methodological point of view, the hybrid tools, for the first time, combine the newly developed parsimonious AHP with a clustering method. From a practical point of view, this paper has introduced a new tool to classify and improve the production quality of the processes to an extremely competitive sector.