文章基本信息

标题：Prediction and Analysis of Seasonal Dynamic Metal Consumption using Aggregated LightGBM - A Case Study
本地全文：下载
作者：Arjun Balamwar ; Rony Mitra ; Manoj K Tiwari 等
期刊名称：IFAC PapersOnLine
印刷版ISSN：2405-8963
出版年度：2022
卷号：55
期号：10
页码：725-730
DOI：10.1016/j.ifacol.2022.09.494
语种：English
出版社：Elsevier
摘要：The steel industry is regarded to be one of the industries that serve many small and large industries. Hence, for a major metal forming industry, understanding the market behavior of the steel consumption and forming is very crucial. Supply and demand lie among the most important features in stimulating the steel market. A critical problem currently facing the industry is finding the optimal inventory. This helps in preventing redundant excess inventory reducing additional cash flow as well as an inventory shortage that leads to an unstable functioning. In most cases, supply and demand have complex functions. Hence finding the optimal inventory requires accurate modeling and forecasting of steel consumption. This paper forecasts the monthly consumption of steel for a prominent metal forming industry in southeast Asia using an aggregated LightGBM method. The model is fed the daily consumption of 35 unique steel for 5 years, from 2014 to 2019 along with multiple critical temporal features. Finally, the predictions of consumption per day are aggregated to a month, which allows the model to get an in-depth analysis of the behavior of each steel consumption.
关键词：Steel Consumption forecasting;Light gradient boosting machine (LightGBM);Temporal features;Dynamic model averaging;Feature engineering;Data driven