A Study on Supplier Evaluation and Selection in Iron and Steel Industry Based on TOPSIS and Random Forest Algorithm

This article aims to construct a supplier selection model for JG Company, a large steel head enterprise, based on a random forest algorithm. The model uses a supplier evaluation indicator system that covers four aspects: quality, price, service, and reputation. The model parameters are determined by experiments on the Tempo big data platform, and the optimal number of trees is set to 20. The model is applied to classify 31 suppliers that have not established cooperative relations with JG Company, and the results show that the model has strong comprehensiveness and high accuracy. The model can be used in two scenarios: one is to dynamically evaluate the existing suppliers and adjust the cooperation strategy according to their levels; the other is to identify potential suppliers in the supply market and provide data support for the next cooperation posture and business negotiation.