Search for a command to run...
With the rapid development of the global economy, international trade plays a crucial role in the economic growth and resource allocation of various countries. As an important logistics channel connecting China and Europe, the China-Europe Railway Express has significantly contributed to supporting the "Belt and Road" initiative and promoting regional economic cooperation. However, the efficiency of container transportation directly affects the quality of transport and operational costs, thus improving the loading efficiency of container transport has become an urgent issue in modern logistics management. Most existing container transport methods rely on traditional scheduling and loading strategies, which often fail to meet the real-time demands posed by dynamic market changes. This paper proposes a new real-time loading optimization framework based on Radio Frequency Identification (RFID) technology to address the container transport optimization problem for the China-Europe Railway Express. This framework manages real-time cargo requests through task queues and dynamically invokes the Iterative Heuristic Tree Search (IHTS) algorithm by the core decision-making component to generate loading plans and pass them to the execution component. By constructing a data generation model based on a normal distribution, this paper simulates the recognition probability of RFID tags to enhance decision-making accuracy. Experimental results show that the proposed method completed 45 tasks within 60 minutes, which is 50.00% higher than the improved Q-learning algorithm and 28.57% higher than the genetic algorithm based on the Metropolis criterion. In terms of path optimization, the length of the path in this method is 108 meters, significantly shorter than the 125 meters of the improved genetic algorithm and the 141 meters of the Q-learning algorithm. In addition, the total transportation cost of the proposed loading optimization method is 608.28 yuan. This cost integrates the vehicle transportation distance cost, the delay penalty caused by failure to load on time, and other operational losses. Experimental results demonstrate that this real-time loading optimization framework not only significantly enhances container loading efficiency but also effectively reduces operational costs, showing promising application prospects and practical value.