Research on Fresh Product Logistics Transportation Scheduling Based on Deep Reinforcement Learning

With the improvement of the economic level, people’s quality of life continues to improve, the demand for fresh food is increasing, and the logistics of fresh products is also developing rapidly. We effectively balance the relationship between transportation costs and service levels in fresh product logistics and transportation businesses, improve the transportation capacity and efficiency of logistics transportation businesses, and improve the resource utilization of businesses. It is important for the development of the logistics transportation scheduling industry for fresh products. Significance. Based on this, this paper proposes a DNQ algorithm based on pointer network, which solves the single fresh product distribution service center-regional efficient logistics scheduling problem, and a feasible logistics transportation scheduling scheme can be obtained through simulation experiments. The simulation results show that the algorithm is superior to other common intelligent algorithms in terms of accuracy and stability, which proves that the algorithm is effective and feasible (the research results cannot be directly shown in the abstract and need to be supplemented) At the same time, it further explored the DNQ algorithm to improve the correction network, which can solve the complex problem of multiple fresh product distribution service centers-regional efficient logistics scheduling. It is a successful attempt to improve the solution algorithm. Complex logistics and transportation scheduling problems provide ideas and have good guidance and reference significance.