Computational model design for road to waterway transportation conversion efficiency improvement based on deep learning algorithm

Path planning for combined transportation can improve the irrationality in the carrier task so as to achieve the purpose of cost reduction and efficiency. This study takes economic efficiency as the starting point, takes the transportation time window as a constraint, and constructs the total cost objective function in the transportation process. The objective function of carbon emission in the transportation process is designed to address the green and low-carbon problem. The Ant colony algorithm is used to solve the multi-objective joint transportation model, aiming to obtain a green and efficient transportation path that meets the actual demand. The ACO algorithm shows good convergence performance in the transportation task from starting point O to ending point D. The ACO algorithm is capable of solving the transportation task from starting point O to ending point D after 130 iterations. After 130 iterations of the algorithm, the cost objective function value converged to 3427.8 RMB, and the carbon emission objective function value converged to 5534 kg after 90 iterations of the algorithm. Furthermore, the ACO algorithm’s Pareto solution set exhibits a uniform spatial distribution and excellent ductility and approximation effects. Assigning the same weights to cost, carbon emissions, and transportation time, the optimal solution 4 is obtained. The scheme from the starting point O via nodes 1, 2, and 6 to the end point D, through the joint transportation of road and waterway for 162 hours, a total of 5549 kg CO2, and the cost of transporting each ton of goods is 4093.9 yuan.