Abstract: Aiming at the time delay optimization problem in multi-cluster tools, a time delay optimization model was established and the solving heuristic algorithm is developed. We first describe the steady state production process of multi-cluster tools. Based on this, the influence mechanism of time delay caused by robots waiting was analyzed, and then a robot waiting time allocation model to reduce time delay was established. An improved Multi-objective evolutionary algorithm based on decomposition-grey wolf optimizer (MOEA/D-GWO) algorithm was proposed to solving the model. Finally, examples were given to demonstrate the efficacy of the methodology.