Research on Supply Quantity Transportation and Ordering Model Based on TOPSIS and Support Vector Regression

Abstract

For the ordering and transportation of raw materials, we should focus on the production efficiency of the enterprise, that is, the purchasing cost of raw materials, the stability of supply, and the loss rate of transportation. Feature engineering is carried out according to the relevant data to generate more supply features, and the evaluation model uses TOPSIS combined with entropy weight to give the supplier ranking. First of all, this paper makes a quantitative analysis of the supply characteristics of suppliers and uses Python to deal with the order quantity and supply quantity data of the suppliers. Based on this, the data mining work is carried out, and the supply characteristics are analyzed quantitatively. And use TOPSIS combined with entropy weight to give the top 50 most important suppliers, and then work out the most economical raw material ordering plan and the least loss transfer plan in the next 24 weeks. Using SPSS software, the descriptive statistics related to the transport loss rate are obtained. Combined with the data analysis of the loss rate, the corresponding ranking of transporters is obtained by using the entropy weight method, and the 0-1 integer programming is constructed for the transshipment scheme. Support vector regression algorithm model is used to predict the most economical raw material ordering plan per week in the next 24 weeks. This paper provides a reference for the ordering and transportation process of raw materials in enterprises.