Under the background of the integration of multiple regulations, there is an urgent need for the coordinated optimization of transportation planning and land use integration in the aspects of urban overall planning, land use overall planning, and comprehensive transportation system planning. In order to achieve this purpose, at the macro level, this paper uses the DEMTEL model to simplify the factors affecting the interaction between the two, and establishes an interactive evaluation model between traffic and land use through non-parametric statistical analysis DEA. Then, the efficiency value of the evaluation result is used as the output value of BP neural network to learn, and the prediction model of traffic planning and land use coordination degree is established. Through the analysis of the prediction results, the key factors can be purposefully adjusted to make necessary preparations for the collaborative optimization of the two. © 2021, Saudi Society for Geosciences.