BMW, like many global manufacturers, is faced with the challenge of scheduling their production runs to achieve production targets while minimizing idle hours. There are a wide range of possible configurations and many constraints. Different shops have different production rates, and each has their own discrete set of shift schedules. What’s more, manufacturers need to prevent overflows and shortages in the buffers between steps in the manufacturing process.
As part of their membership in MIT’s The Center for Quantum Engineering (CQE), Zapata AI and BMW Group collaborated to apply generative AI techniques to BMW’s plant scheduling optimization problem. Specifically, we trained a quantum-inspired generative model on the best solutions generated by existing state-of-the-art solvers. The generative model then generated new, previously unconsidered solutions — tying or outperforming other state-of-the-art solvers in 71% of problem configurations.
Model computational fluid dynamics for new designs with enhanced accuracy.
Build design process optimization models capable of balancing multiple objectives for product functionalities with safety, reliability and cost.
Optimize the scheduling of machine processes and employee shifts.
Optimize the timing of automated processes on the factory floor.
Proactively predict when machines will need maintenance.
Find the optimal satellite configuration to maximize coverage and signal quality while minimizing operational costs by leveraging quantum or quantum-inspired prescriptive analytics.
Optimize transportation networks to reduce costs and delivery times and select suppliers and vendors optimized for product quality and demand coverage.
Optimize distribution routes to reduce fuel costs and delivery times.
Optimize the stocking of raw materials and components for manufacturing and for the stocking of finished products in distribution warehouses.
