Designing a New Dry Port-Seaport Logistics Network with a Focus on Industry 5.0 by Machine Learning

Dry Ports (DPs) are essential inland terminals that cover a variety of goods and are strategically established in several countries. According to the general structure of DPs across various countries, they play a major role in the field of multimodal transportation. Considering that some countries have wide beaches, they can earn non-oil income by using these beaches. Therefore, in this study, we are looking at the design of the Logistics Network (LN) for Dry Port-Seaports (DPSs), with an emphasis on multimodal transportation systems, by predicting demand utilizing a Machine Learning (ML) algorithm named Seasonal Autoregressive Integrated Moving Average (SARIMA). The mathematical model of this study includes three Objective Functions (OFs) (e.g., minimization of cost and environmental impact, and maximization of job opportunity and resilience). This work offers a mixed-integer programming model for setting up a viable LN by the Industry 5.0 (I5.0) dimension. Moreover, the Lp-metric, Goal Programming (GP), and Meta-GP (MGP) approaches are used to solve the developed model. Eventually, several quantitative illustrations are employed to evaluate the mentioned solution approach.