ABSTRACTS

摘要

Association rule mining is originally developed for the market basket analysis but expected to be used for a wide range of data analysis such as fault analysis in manufacturing. However, in real plant operation, only batch-level fault data is available. Thus, without unit-level data, the straight forward application of basket analysis is difficult. In this paper, we propose a way to extract a frequent fault pattern from the aggregated dataset. We show how to generate a virtual basket from aggregated basket. The fault pattern analysis in a process industry is also shown as a real example.