Introduction: This study aims to develop a hub (Distribution Centers) location model for supply chains of seasonal perishable goods under a multi-modal transportation system. Due to their inherent characteristics, perishable products are highly susceptible to uncertainties & disruptions, & even minor delays or interruptions at any stage of the supply chain can lead to quality degradation, lost sales, & risks to public health. One major source of uncertainty in such supply chains is the seasonal variability of product demand. Moreover, distribution centers play a critical role in fulfilling the demand for seasonal goods. Therefore, making decisions regarding hub location (construction or rental) can be an effective strategy for responding to dynamic demand in these supply chains.
Methods: A bi-objective, multi-product, multi-modal mathematical model is developed in this article. The model aims to minimize various supply chain costs, including hub location costs, transportation costs, shortage costs, & product deterioration costs during transportation, as well as to minimize product transfer time. The proposed problem is formulated as a mixed-integer programming model. To solve the bi-objective problem, exact solution methods, normalization, & augmented e-constraint method approach are implemented in the GAMS using the CPLEX solver.
Results and discussion: Furthermore, numerical experiments of varying scales are conducted to validate the model & assess its scalability Moreover, increasing the parameter associated with the maximum number of rentable hubs leads to a reduction in total costs as well as shortage levels across different seasons. Therefore, during periods of temporary demand surges, seasonal rental of hubs instead of permanent establishment can significantly reduce both total costs and shortage levels. However, this approach simultaneously results in an increase in the total number of established and rented hubs, which consequently enlarges the average distance between hubs and demand points, leading to longer travel times. This outcome highlights a fundamental trade-off between the overall network cost and travel time, or equivalently, the service level in the supply chain of perishable products. Furthermore, the utilization of multiple transportation modes (air and road) in this problem, each with its distinct characteristics, plays a crucial role in improving product distribution. For instance, considering the perishable nature of the products, the use of air transportation enhances delivery speed, while road transportation, due to its high accessibility, enables coverage of most regions and facilitates product distribution.
Conclusions: By integrating bi-objective modeling, demand dynamics, & multi-modal transportation, this study provides an effective framework for managing dynamic demand for perishable goods. The results can assist supply chain managers & policymakers in making strategic decisions such as hub location in industries with seasonal products or services, including tourism to ensure faster & more efficient supply during peak seasons while reducing costs in low-demand periods.