Abstract:The industrial sector is the main source of energy consumption and environmental pollution, and effective investment in the industrial sector has an important impact on economic development to a certain extent. In orderto implement the concept of green development and promote energy conservation and emission reduction, a study on the effectiveness of investment in green industry was proposed. Based on the undesirable SBM-SVM model and its improvement, the relevant indicators of industrial enterprises in Chongqing from 2011 to 2020 were selected as the sample data, and the evaluation efficiency obtained by the undesirable SBM model was divided into two categories: effective investment and ineffective investment as the outcome variables. The input and output indicators were used as characteristic variables to construct an SVM model to study the classification and prediction of industrial investment effectiveness. Through trial-and- error method, PSO, and GA intelligent optimization algorithms, the penalty factor C and kernel function parameter g of the SVM model were optimized. The results showed that the optimization effect of the PSO method was the best, and the accuracy rate was increased from 71.88% to 88.66%. The constructed new undesirable SBM-SVM model has certain feasibility and applicability to classify the effectiveness of industrial investment after improvement and optimization.