Research on Genetic Algorithms to Plan the Ordering and Transportation of Raw Materials in Enterprises

Abstract

Listen

Translate article

To ensure the efficient and economical production and operation of enterprises, overall arrangements for suppliers and transporters are an adequate basis for businesses to run enterprises. According to the data research scheme given in this paper, it is used through the mathematical software MATLAB, SPSS, planning analysis, entropy method etc., which adopts genetic algorithms. The problems of selecting merchants, ordering schemes, and transhipment schemes are modelled and analysed. For the first question, the entropy weight method is used, which selects 50 merchants. It is constructed a data matrix and evaluates the 50 merchants. Normalization is carried out to obtain positive and negative indicators, which are defined and functionally expressed. The proportion and entropy of 402 evaluation objects in this index are calculated for the five indexes. Then the difference coefficient is calculated by entropy value. Furthermore, the total score of the evaluation object is calculated. The 402 evaluated objects were sorted, and the top 50 with the highest score were selected. The second question is divided into three small questions. Based on the data of 402 suppliers, the first question predicts the future 24 through time series. The weekly supply is studied by establishing decision variables x1, x2, etc., in which x1 can take 0 or 1. Here, 1 means to choose this supplier, and 0 means not to choose. A 0–1 programming model is built with the minimum sum of all suppliers as the objective function. The constraint conditions of A, B, and C suppliers can provide the raw materials A, B, and C divided by 0.6, 0.66, and 0.72, respectively, and it is added up to equal to 28200 cubic meters. A genetic algorithm obtains 127 suppliers, and the second question is on the premise of 127 suppliers. The objective is to minimize the economic sum. The number, decision variables, and constraints are unchanged, and a genetic algorithm obtains the optimal scheme. The third question takes the choice of suppliers and forwarders as the decision variables, which is still a 0–1 programming model and takes the minimum loss rate as the objective function. The total weekly supply of suppliers is less than 6000 and one. The supplier can only choose one forwarder as the constraint condition, and a genetic algorithm can obtain the transhipment scheme. In view of the third question, it is assumed that the weight of raw material A is 100, and the weight of raw material C is 1. The objective functions are the minimum sum of supply and the minimum loss rate after weighting. The decision variable is still the choice between the forwarder and the supplier, and the constraint condition is that the sum of supply is less than 6000 cubic meters. A genetic algorithm obtains the transport scheme. For the fourth question, it is greatly improved and the cost is not limited because of the production capacity of the enterprise. The constraint conditions and decision variables are the same as those of the third question, and the objective function is the same as that of the second question. A genetic algorithm obtains the most suitable transfer scheme.