Campus intelligent water supply system is an important part of modern campus facilities, but through comprehensive statistical analysis of campus water supply system, it is found that there is water leakage in campus water supply system. Therefore, this paper establishes an average water loss rate model to find and judge the water leakage of pipelines in time. Then, according to the residual diagram and the abnormal data of correlation analysis, the leakage water meter is determined. Finally, in order to ensure the lowest maintenance cost, an integer programming model of 0- 1 is established, and the optimal maintenance decision scheme for pipeline network maintenance is designed.