Spatial-temporal out-of-order execution for advanced planning and scheduling in cyber-physical factories

Cyber-Physical System (CPS) is one of the most promising directions of Industry 4.0 smart manufacturing. Abundant manufacturing data and information are available for decision-makers in real-time thanks to the application of various frontier technologies in CPS. However, the inherent complexity and uncertainty of manufacturing optimization still plague scholars and practitioners and impede further progress of smart manufacturing. The production planning and scheduling is such a complex and stochastic problem that has received considerable research attention. Whereas how to leverage the strengths of CPS for breaking the bottleneck of complexity and uncertainty, is still a question that needs further exploration. This paper proposes a novel divide and conquer approach, Spatial-Temporal Out-Of-Order execution (ST-OOO), for achieving real-time planning and scheduling in cyber-physical factories. ST-OOO divides the space and time scopes of a factory into finite areas and intervals to reduce complexity and localize uncertainties so that the original complex optimization problem is decomposed into a set of subproblems with different spatial and temporal characteristics. These small-size subproblems can be assembled using data and information visibility and traceability, and then solved in a rolling spatiotemporal manner to generate a global solution. A case study shows that ST-OOO has a well-balanced and more stable performance compared to traditional strategies. Sensitivity analysis is carried out to study the impacts of spatial and temporal scales on the results.

Automated summary

Cyber-physical systems (CPS) hold great potential in smart manufacturing, but the complexity and uncertainty of manufacturing optimization remain a challenge. This paper introduces a novel divide and conquer approach, Spatial-Temporal Out-Of-Order execution (ST-OOO), to decompose the complex optimization problem into smaller subproblems and generate a global solution through rolling spatiotemporal execution.