Production Planning Forecasting System Based on M5P Algorithms and Master Data in Manufacturing Processes

With the increasing adoption of smart factories in manufacturing sites, a large amount raw data is being generated from manufacturers’ sensors and Internet Things devices. In environment, collection reliable has become an important issue. When utilizing collected or establishing production plans based on user-defined data, actual performance may differ established plan. This particularly so when there are modifications physical line, such as manual processes, newly developed addition new equipment. Hence, reliability current cannot be ensured. The complex characteristics manufacturers hinder prediction future existing data. To minimize this problem, M5P algorithm, used to predict dynamic using baseline information that can predicted. It combines linear regression decision-tree-supervised machine learning algorithms. algorithm recommends means reflect predicted plan provides results compared with information. By comparing planning changed master it help management determine impact work time quantity confirm plans. forecasting directly affects capacity resources, well times schedules, ensure efficient planning.