Data Mining in Manufacturing: A systematic literature review identifying the current trends and future research direction

In modern manufacturing environments, large amounts of data are collected from database management systems and data warehouses from all engaged areas, such as fault detection, maintenance, operations, product quality development, and production process. This paper reviews the literature in the broad manufacturing domain dealing with knowledge discovery and data mining applications with special emphasis on the type of manufacturing functions to be performed on the data. Data mining in manufacturing has attracted a great deal of attention in recent literature but only a very few systematic and extensive research reviews capturing the dynamic nature of this topic. The rapidly growing interest in applying data mining in manufacturing from both academics and practitioners has urged the need for a review of current research and development to create a new agenda. Selected 75 papers were classified into five research categories namely conceptual and empirical papers on fault detection, maintenance, operations, product quality development, and production process. The review primarily attempted to achieve the following two research objectives: (1) to conduct a systematic literature review on applications of data mining techniques and algorithms in manufacturing and (2) to compile and critically evaluate the existing gaps in the literature and real-world manufacturing and identify future directions in the research apart from the critical analysis of the literature. Finally, the scope of future research direction was discussed in detail on sustainable manufacturing, agile manufacturing, lean manufacturing, intelligent manufacturing, and manufacturing in times of pandemic, respectively. Keywords: Data mining, Process Mining, Fault Detection, Maintenance, Operations, Product Quality Development, Production Proces