The aim of this study is to investigate the feasibility of partially replacing cement with eggshell powder to prepare green and low-carbon foam concrete, and to propose a method to predict the performance of eggshell powder foam concrete. In this study, the laboratory experiments were first conducted to analyze the influences of eggshell powder content on water absorption, mechanical strength, and pore structure of foam concrete. Subsequently, based on the experimental results, a model to predict the performance of eggshell powder foam concrete was proposed using the machine learning approach. Results indicate that foam concrete specimens mixed with eggshell powder exhibit reduced initial surface absorption rate and water absorption rate. Additionally, the addition of eggshell powder can improve the compressive and flexural strength of foam concrete, with the highest strength observed at a 15% substitution rate. Furthermore, the increase in foam volume leads to pore size expansion, with large pores accounting for a greater proportion of the foam volume as a result of eggshell powder addition. Notably, the filler effect of eggshell powder is more pronounced, with an addition of 15% eggshell powder leading to an increase in the proportion of small and large pores in the foam concrete. The machine learning model developed in this study demonstrates its effectiveness in predicting the performance of foam concrete containing eggshell powder. When compared with the decision tree model, the prediction results of the random forest model exhibit higher accuracy.