引入神经网络的应急调度系统设计

Design of emergency scheduling system introducing neural network

当前的应急调度系统在进行物资调度的过程中,遇到需求点分散、路况通行不畅、时间紧急等问题,调度智能性很低,无法满足应急需要。在此设计并实现一种引入神经网络算法的应急调度系统,根据需要应急调度的地区、交通状况、需求紧张程度等参数构造神经网络模型,以调度算法的最短路径为网络输出,以微控制器为软件算法信息处理核心,建立硬件模块。以某物流公司的应急调度系统为测试对象,结果表明,新系统在引入神经网络算法后,能够塑造可靠性较高的应急调度模型,具有很强的实用性。

Since the current emergency dispatch system in the process of the material scheduling has very low scheduling intellectuality and can not meet emergency demands when it faces with the cases of demand point dispersion,poor road traffic and time urgency,an emergency dispatch system with neural network algorithm was designed and realized. The neural network model was built according to the parameters of emergency dispatch area,traffic condition and demand tension degree. The hardware module is established by taking the shortest path of the scheduling algorithm as network output and the micro controller as the information processing core of software algorithm. The emergency dispatch system in a logistics company is taken as test object. The test results show that the new system can shape the emergency scheduling model with high reliability after introducing neural network algorithm,and has very strong practicability.