This research uses apriori algorithm and FP-growth to discover association rules mining from maintenance transaction log of ATM maintenance. We use ATM maintenance log data file from year 2013 to 2016. In pre-process step, we clean and transform data to symptom failure part attribute. Then, we focus on comparison of association rules between FP-growth and apriori algorithm. The result represents that FP-growth has better execution time than apriori algorithm. Additionally, the result from this paper helps maintenance team to predict symptom of failure or failure parts in future. As the advantage of predict failure parts, maintenance team will prepare a spare parts in store and prevent break down time of machine. The team can add failure parts from rules to preventive maintenance to prevent fail machine.
CITATION STYLE
Rachburee, N., Arunrerk, J., & Punlumjeak, W. (2017). Failure part mining using an association rules mining by FP-growth and Apriori Algorithms: Case of ATM maintenance in Thailand. In Lecture Notes in Electrical Engineering (Vol. 449, pp. 19–26). Springer Verlag. https://doi.org/10.1007/978-981-10-6451-7_3
Mendeley helps you to discover research relevant for your work.