engleski

Application of data mining techniques in small- series job shop

Manufacturing industry has its domain where fast and reliable decisionmaking is often an exigency. Modern ERP, MRPII and MDC systems accumulate large amounts of data that are stored in databases located onsite or in cloud- based servers. Although these data are valuable for the company as they are, they hide even greater potential value. A large amount of data increases the time required to conclude or make decisions based on that data. Regarding that fact, it can be concluded that the potential value of the data is in the efficient analysis and interpretation which can help in the decision-making process. Nowadays, job shops are an important part of the industry in Croatia, but their decision making process mostly relies on skilled employees whose knowledge and possibilities, faced with high amount of available data, is getting more and more inadequate. This paper considers machine learning algorithms application for data mining in small-series job shop. Some of the data mining techniques are reviewed, applied and evaluated on the manufacturing data with the aim of demonstrating the potential of improving the decision-making process in the job shops using RapidMiner tool.

Data mining; machine learning; job shop

nije evidentirano