|
چکیده
|
In perishable seasonal products supply chains, it is critical for managers to make appropriate and timely facility-related decisions during the planning horizon; because according to the dynamic nature of product transportation demand, it is necessary to coordinate the network configuration and operational capacities of the facilities due to variations in demand over time. This paper presents a non linear programming model for a dynamic and sustainable hub location problem with ramp-type and time-dependent transportation demand. The model’s objectives are the minimization of total system cost (i.e., costs associated with transportation, opening and closing hubs, adjusting hubs’ capacity through modules, and CO2 emissions) and the maximization of responsiveness (through maximizing the total minimum product quality in all time periods) simultaneously. To solve the model, two meta-heuristic algorithms, namely NSGAII and SPEAII, are presented. The proposed model is validated using three well-known datasets (i.e., the CAB, AP, and TR datasets) and a test problem on the Turkish network for freight transportation (based on a mixed-integer programming (MIP) model that results from fixing the breakpoints in the initial model). Also, the results of some sensitivity analyses and statistical tests are presented. The findings show that the results of the proposed dynamic model in terms of the number of Pareto solutions and some multi-objective indexes (i.e., quality metric, mean ideal distance, and spacing metric) is better than its equivalent static model. In terms of some multi-objective indexes (i.e., mean ideal distance and spacing metric), the NSGAII outperforms the SPEAII. Also, the SPEAII outperforms the NSGAII in terms of diversification metric and solution time. Additionally, the results show that if the economic and responsiveness objectives are considered simultaneously, the cost of network design is higher than when only the economic objective is considered.
|