Abstract: The traditional production management and control methods based on regular production planning is no longer applicable in the multi-varieties and small-batch production mode due to the highly dynamics caused by random orders, service resources, dynamic cooperation and other factors. In Industry 4.0 and other intelligent production mode, how to obtain global system interference, dynamically determine process control needs and iterative optimization control objectives in order to guide the production system to continuously adjust the execution plan and configuration structure to keep the production system in an optimal state under dynamics adaptively is an important challenge for the production system management layer. A large-scale production system consisting of multiple production units (one workshop for example) was selected as a general research object and the dynamic operating environment is fully defined with changes in external orders, self-adjust control and the underlying execution interference. In addition, a three-state conducting, multi-stage synchronized controlling optimal-state operational control framework is proposed in this study, which described as to get the optimal guide path real timely through the dynamic matching mechanism of "theoretical optimal state-actual good state-actual state", and then get the local optimal state control under multi-stage synchronized control with Analysis Target Cascading method to ensure the system under an optimal target path for optimal operation. Finally, combining the production-logistics and storage optimal synchronization operation scheme designed by the research team for a cooperative enterprise, it provides a feasible framework and method for the production system to face the high dynamic problem.

Key words: collaborative decision-making, dynamics, internet of things, optimal state, synchronized control