Understanding production chain business process using process mining: A case study in the manufacturing scenario

Abstract

Due to the continuous market change the enterprises need to react fast. To do that a better understanding of the way to work is needed. Indeed this was a real need of a manufacturing enterprise working in the production of coffee machines and selling them all over the world. In this paper, we present the experience made in the application of process mining techniques on a rich set of data that such enterprise collected during the last six years. We compare five mining algorithms, such as: α-algorithm, Heuristics Miner, Integer Linear Programming Miner, Inductive Miner, Evolutionary Tree Miner. We evaluated algorithms according to specific quality criteria: fitness, precision, generalization and simplicity. Even if comparison studies are already available in the literature we check them according to our working context. We conclude that the Inductive Miner algorithm is especially suited for discovering production chain processes in the context under study. The application of process mining gives the enterprise a comprehensive picture of the internal process organization. Resulting models were used by the company with successful results to motivate the discussion on the need of developing a flexible production chain.