In this paper a knowledge based assembly control system for automobile manufacturing is presented which has been implemented for the B M W manufacturing facilities in Munich. The system is fully integrated in the process control system. Its main purpose is to support the assembly supervisor in taking appropriate corrective actions in case of abnormal process conditions. The distinctive feature of the system is its capability to validate actions proposed by a rule based troubleshooting component through a one to one simulation model of the real manufacturing process. Thus effectiveness of actions at the first attempt is ensured and predictive process control is supported. The approach proposed in this paper can be generalized to a domain independent scheme for manufacturing process control.