Abstract: In response to the problems of low efficiency, low accuracy, and lack of scientific evaluation system in current evaluation methods, an evaluation method combining BP neural network and Analytic Hierarchy Process was designed. In the specific evaluation process, it was achieved by constructing an evaluation system, calculating indicator weights, creating a network structure, selecting training samples, and verifying the model. Firstly,the paper analyzed the landscape design under the concept of low impact development, identified existing problems, and constructed an evaluation system. Then, Analytic Hierarchy Process was used to construct a hierarchical model and determine the weights of each indicator. Then, construct a BP neural network model was constracted and trained with sample data. Finally, select 10 samples were selected from 100 samples for simulation testing, and applied them to actual cases. The results show that this method had higher evaluation accuracy compared to traditional single Analytic Hierarchy Process and BP neural network evaluation models. In case applications, the test results were consistent with expert evaluation results, making the evaluation effect of landscape design under the low impact development concept ideal. It can effectively adapt to the complexity and diversity of modern landscape design and provide scientific reference opinions for relevant practitioners.