The construction process of cutter suction dredger is very complex，with more uncertainty and randomness. Regarding the problems of many impacts in dredging output，insufficient stability and low productivity，we take the slurry in the transmission pipeline of cutter suction dredger as the research object，analyze the characteristics and influencing factors of dredging construction，preprocess the historical big data of dredging engineering，compare two different flow prediction schemes of T-S model and historical data nearest neighbor，and use deviation feedback controller to independently optimize construction parameters based on big data. The results show that the accuracy of the second scheme is higher because the data are preprocessed. The flow adjusted by feedback control according to the optimization parameters is higher than the original flow.