Abstract: Because the sensors distributed on the bridge continuously collect data for a long time and transmit a large amount of data, the traditional bridge health monitoring system has not been able to fully deal with the massive monitoring data. Therefore, this paper proposed a sliding window data flow processing model driven by K-line chart time slice, which could quickly and effectively collect the data flow in the sensor network and reduce the data storage for bridge monitoring.The cable force statistical evaluation system based on a single server node has a computing bottleneck in the analysis of massive data. Using the distributed parallel computing of cluster, the paper proposed a cable force parallel processing model based on Map/Reduce. The experimental results show that the processing of cable force monitoring history big data under this model can significantly reduce the analysis and evaluation time of expert system.