Algorithms And System Implementation Of RFID-enabled Real-time Vehicle Scheduling

Along with the diverse developments of form of distribution and the rapid expansion of its scale in modern logistics, traditional vehicle scheduling commonly has difficulties in meeting the requirements in e-commerce environment such real time, efficiency, low-cost. Improvement and optimization of vehicle scheduling is one of the important ways to improve the efficiency of modern logistics distribution, to cut transportation costs, and to satisfy customers. So any in-depth study on the vehicle scheduling problem has theoretical and practical significance.Most experts and scholars working on the vehicle scheduling problem are mainly at the stage of non-real-time scheduling. Such a problem is also called static scheduling problem in that the relevant data in planning vehicle are known in advance and assumed to be constant. However, when RFID technology is widely applied in modern logistics distribution systems for real-time tracking and management of vehicles and goods, the sensored data are uncertain and varying in real-time. It is difficult for non-real-time scheduling to deal with such dynamics, whereas real-time vehicle scheduling can work on real-time data to plan vehicle routes acceptably.In this thesis, we propose an RFID-enabled real-time scheduling framework to solve real-time vehicle scheduling problem with uncertain available vehicles. The framework incorporates management of real-time RFID data, scheduling real-time tasks and coordinating logistics system modules to achieve collaborative working. To solve the vehicle routing problem, we change the encoding of the particle swarm optimization and correct non-standard particles through re-initializing the particle, thus propose a particle swarm optimization algorithm based on the standard population.On the basis of the real-time scheduling framework, we finally implement an RFID-enabled decision-making system in Java. Experiments demonstrate that each module is real and valid, and the system performs well and can effectively improve the efficiency of distribution in logistics.