Research On Multi-Period Location-Inventory-Route Problem For Auto Parts Supply Logistics Network

Under the third-party logistics operation mode,it is a top-down decision-making problem for auto parts supply logistics network to integrating consider location layout,inventory control and route planning,which has critical influence on the integrated operation of logistics network.So far,there exists several practical problems in the integration optimization of auto parts supply network,such as low integration degree of associated business,insufficient consideration of multi-period inventory changes,and difficulty in coordinating multi-period demand of production line.Therefore,it is urgent to deeply discuss joint decision on the multi-period location-inventory-route(MLIRP)of the auto parts supply network,so as to reduce the system operation cost and carbon emisson,which is beneficial in improving the core competitiveness of the auto supply chain.Based on the previous research results of location-inventory-route model(LIRP),this paper presents a combined MLIRP problem in auto parts supply logistics network.Regarding the location of the Centralized Distribution Center(CDC),multi-cycle and multi-product inventory decision,and routing planning as core content,this paper propose a Mixed Integer Non-linear Programming(MINLP)model to solve the joint MLIRP decision problem,so as to minimize total system cost and carbon emissions.Furtherly,considering the uncertainty of Automobile Production Line(APL)demand for parts caused by market fluctuations,a robust optimization model of MLIRP with uncertain demand is constructed.Aiming at the of computional complexity of the proposed model,this paper utilizes GUROBI solver for smallscale cases and an improved artificial bee colony algorithm based on GUROBI for large scale experiments.In order to test the validity of the model and the rationality of the algorithm,this paper integrate the basic data of Changchun auto parts logistics to design examples and analyze the results.Through analysis,the main research results of the paper demonstrate that:Firstly,in the dual-objective optimization,the total system cost of the auto parts supply logistic network is more sensitive to cost parameters involved in the model.However,the carbon emission is only sensitive to the inventory holding cost and the CDC collecting cost.Secondly,under the circumstance of uncertain demand,with the increase of conservative degree,the inventory capacity of the CDC gradually increases and auto part suppliers tend to choose the mode of direct delivery to APL.Thirdly,the results demonstrated that GUROBI solver can quickly obtain the optimal construction scheme of small-scale parts supply logistics network.Moreover,for large-scale experiments,the improved artificial bee colony algorithm based on GUROBI is able to improve the efficiency.and the fast search ability.In conclusion,the proposed MLIRP joint decision model and solution algorithm give full consideration to the system operation cost control and carbon reduction management,which provide a guidance for decision-makers to effectively solve the multi-period locationinventory-route problem of automotive parts supply logistics network.