In order to study the distribution law of ground settlement caused by urban subway tunnel construction and do a good job in forecasting and pre-control, based on big data analysis thinking, the applicability of Peck formula was analyzed by using settlement data of tunnel excavation section in advance. On the premise that the applicability was verified, a three-dimensional spatial distribution model of Peck formula with uncertain parameters was established. The uncertain parameters of the model were transformed by function, and the deterministic parameters were obtained by inverse transformation of the solution according to the function transformation conditions used the least square method. A complete prediction model of tunnel construction surface settlement can be obtained by substituting Peck formula. Taking the monitoring data of Qingdao subway tunnel construction process as an example, the settlement prediction model obtained by the above method was used to calculate the predicted settlement value of the monitoring point, and the posterior difference test method in the grey prediction theory was used to test and evaluate the accuracy of the surface settlement prediction model. The evaluation results show that the qualified rate of fitting accuracy between the settlement value obtained by the model and the measured data is 100%. It shows that the parameters and model of this method are reasonable and feasible, and can effectively predict the surface settlement value during tunnel construction.