复杂供应链网络资源传输数据自动调度方法

为了解决传统方法实时性较差、公平性极低、容易造成复杂供应链网络崩溃,以及无法有效调度拥塞情况下数据,导致调度性能下降的问题。研究了一种新的复杂供应链网络资源传输数据自动调度方法。分析了传输数据过程中出现数据包乱序现象的原因,发现在复杂供应链网络出现拥塞的情况下,拥塞部分和网络路由部分会出现很大的改变。针对拥塞部分进行调度时,对数据传输过程中拥塞状态下数据调度问题进行描述。结合公平性考虑,通过粒子群算法进行求解。针对路由部分进行调度时,考虑带宽预测与前向传输时延。对各子路径接收端成功接收数据包数量进行预测,依据子路径往返时延,完成对子路径的排序,按照顺序调度数据传输。在出现丢包情况时,通过最大ssthresh优先法对丢包重传路径进行选择。结果表明:所提方法可动态自适应调度不同路径的传输数据,分配比率相对稳定;接收缓存空闲空间变化较少,乱序数据包数量明显降低,拥塞窗口数量最低;公平性因子值明显低于其他方法。可见所提方法调度性能优,可达到复杂供应链网络资源传输数据调度要求。

Automatic Scheduling Method for Resource Transfer Data in Complex Supply Chain Networks

In order to solve the problem of poor real-time performance and fairness of traditional methods, it is easy to cause the collapse of complex supply chain network, and the data can not be effectively dispatched under congestion, resulting in poor scheduling performance. A new automatic scheduling method for resource transmission data in complex supply chain networks is studied. The causes of data disorder in the process of data transmission are analyzed. It is found that the congestion part and network routing part will change greatly in the case of congestion in complex supply chain network. When the congestion part is scheduled, the data scheduling problem under congestion state in data transmission process is described. Considering fairness, the problem is solved by particle swarm optimization (PSO). Bandwidth prediction and forward transmission delay are considered when scheduling the routing part. The number of packets received by each subpath receiver is predicted. According to the round-trip delay of the subpath, the subpath is sorted and the data transmission is scheduled in sequence. In case of packet loss, the maximum ssthresh priority method was used to select the retransmit path. The results show that the proposed method can dynamically and adaptively schedule transmission data of different paths, and the allocation ratio is relatively stable; the free space of receiving buffer changes less, the number of disordered data packets decreases significantly, and the number of congestion windows is the lowest; the fairness factor value is significantly lower than that of other methods. It is concluded that the scheduling performance of the proposed method is excellent, and it can meet the requirements of resource transmission data scheduling in complex supply chain networks.