Composite strata is a common type of strata during the process of shield tunnelling. However,there is relatively little research on shield tunneling parameters in composite strata. In order to study the correlation between different shield parameters, in this paper, based on the Changsha Rail Transit Line 6 Hubai interval tunnel project, the shield tunnelling parameter prediction model was constructed using the random forest (RF) algorithm. The shield tunnelling parameters was predicted using the PSO-RF hybrid algorithm model combining the mathematical statistics and the RF. And the prediction results were verified for errors. The results of the study show that the stratigraphic parameters of the shield tunnel intervals are combined with the shield tunneling parameters. Two types of parameters are used as input and output parameters, which are imported into the PSO-RF hybrid algorithm model for training and learning, 5 predicted shield tunneling parameters are obtained. The mean absolute percentage error between the predicted and actual values of those parameters is generally less than 20%, which indicate that the prediction results meet the requirements of generalisation and prediction ability.