Huffman material purchasing decision-making algorithm

Abstract

The invention provides a Huffman material purchasing decision-making algorithm. Firstly, a purchasing decision-making tree is established by utilization of a C4.5 algorithm; then, the inherit property amounts, namely the weights, of all leaf nodes of the decision-making tree are calculated; the decision-making value of a link from a certain leaf node to a root node of the decision-making tree is obtained through the weights; the weights are established into a Huffman tree by utilization of a Huffman algorithm; leaf nodes of the Huffman tree are output by utilization of a postorder traversal method; and then, priority sorting of purchasing decision-making schemes is obtained. By means of the Huffman material purchasing decision-making algorithm provided by the invention, a purchasing analysis framework, which is easy to understand and relatively high in accuracy rate, can be obtained; the property inheritance values of the leaf nodes of the decision-making tree are calculated through information gain in a mode of simulating children to inherit parent's properties; the higher the inheritance value is, the higher the value of the link from the leaf node to the root node of the decision-making tree is; the leaf nodes are constructed into an optimal binary tree (the Huffman tree) through the Huffman algorithm; the leaf nodes of the Huffman tree are output by utilization of the postorder traversal method; and thus, decision-making analysis is intuitional and ordered.

Description

technical field [0001] The invention relates to the field of enterprise management, in particular to using algorithms to analyze the problem of manufacturing material procurement. Background technique [0002] With the integration of the global market and the advent of the information age, professional production can play a huge role, and the proportion of corporate procurement has also increased greatly, and the importance of procurement has been increasingly recognized by people. Globally, in the product composition of industrial enterprises, the cost of purchased raw materials and parts varies with different industries, generally ranging from 30% to 90%, with an average level of more than 60%. From a global perspective, for a typical enterprise, procurement costs (including raw materials and components) account for 60%. In China's industrial enterprises, the procurement cost of various materials accounts for 70% of the enterprise's sales cost. Obviously, procurement cos...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is in order to overcome the problem of existing algorithm: the decision tree information amount that is constructed by C4.5 algorithm is many and more complex, is not easy to obtain an exact solution

Images