Open AccessSales Forecast of CNC Machine Tool Suppliers Based on Ensemble Learning
Author(s) -
Kai Zhong,
Jiwen Wen,
Juanjuan Zheng
Publication year - 2019
Publication title -
journal of physics. conference series
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.21
H-Index - 85
eISSN - 1742-6596
pISSN - 1742-6588
DOI - 10.1088/1742-6596/1314/1/012129
Subject(s) - random forest , generalization , computer science , process (computing) , product (mathematics) , ensemble learning , industrial engineering , machine learning , manufacturing engineering , data mining , engineering , mathematics , mathematical analysis , geometry , operating system
CNC machine tools, whose product sales forecast is a complicated dynamic process, are the basic equipments for manufacturing industry. It is difficult for a single prediction model to achieve the expected effects because product sales can be affected by multiple factors with features like time-varying, non-linear and random. This paper, which takes the sales data of a CNC machine tool supplier as an example, designs an ensemble learning model based on XGboost and random forest, obtaining lower prediction error through data cleaning, data conversion and parameter optimization. The experimental results show that the generalization ability of this model is excellent, so it can provide decision supports for CNC machine tool suppliers in purchase and sales plans.