Optimization Research Of Purchasing Decision Considering Multiple Transport Alternatives

During the whole manufacturing and operation period, as the starting point of logistics and fund flow of an enterprise, procurement plays a significant part in the latter parts of the business activities. With the acceleration of economic globalization and thorough development of supply chain, the procurement segment itself as well as the role it plays has gradually made a change, and more and more enterprises regard optimization of procurement decisions as the origin of force and strong support of escaping from financial crisis. As with procurement itself, with the rapid development of global procurement and the third party logistics, optimization of transportation exerts an important influence on procurement performance. It is of great practical significance to take the choice of multiple transport alternatives into consideration during procurement decisionsIn this thesis, preliminary research on production procurement and transport alternatives has been done. More Specifically, the author at first makes statistics on literature of supplier choice and procurement fields in the recent ten years. By the analysis of Pareto diagram, the present critical factors that influence procurement decisions are obtained. Secondly, we respectively make a discussion about the effect of choice of multi-transport alternatives on procurement decisions, afterwards, a multi-objective mixed integer programming model considering multi-transport alternatives that aim at two cases of a single product and multi-product is made, and the goal of this model is to minimize total cost, for the sake of accelerating evolution and ensure convergence, self-adaptive mechanism is introduced into genetic parameters, delayed delivery rates and defect rates. Then, appropriate algorithm is chosen for the model and improvements are made for encoding modes and genetic operators in order to increase the efficiency of solving. At last, a numerical example is used to verify the validity of the model and algorithm, managerial analysis and interpretation are made from the results.In this model, we first introduce the concept of a deductible value to handle the problem of purchases higher than actual purchases, the present critical factors that influence procurement decisions are obtained and increase the applicability of the model. At second, analysis is made and expressions of the various elements of procurement decisions after introducing transport alternatives are given. Finally, the transport business is divided into entire transport and zero transport according to actual situation. Based on their characteristics, we come to the simulation of total cost, the determination of delayed delivery rate and the method of calculation of total defect rate. In algorithm, we simplify the encoding modes. And on the basis of the characteristics of the gene value of none-uniform features, crossover and mutation operators are specially designed to ensure that the new individual is still capable of meeting the value demand of the gene. In addition, we come to the simulation of total cost, the determination of delayed delivery rate and the method of calculation of total defect rate, for the sake of accelerating evolution and ensure convergence, self-adaptive mechanism is introduced into genetic parameters, and provisions for treatment of constraint conditions are made.By using this model and algorithm, we handle the coordination decision-making problem of procurement and transport in a better way. On the one hand, we can obtain a series of Pareto solution set needless of future knowledge, which enables procurement decision makers make decisions on the basis of having some understanding for the value and distribution of each objectives of procurement, thus avoiding the subjectivity and blindness of the decision making. On the other hand, the parameters of the model is objective value and relatively simple, therefore, decision makers can use the last arrival as the input parameters of the next order in the practical application, which reflects the rapid response characteristics of the agile supply chain.