Production scheduling optimization is the key to improve production efficiency and on-time delivery rate in the field of intelligent manufacturing. Aiming at the problem of production scheduling in intelligent manufacturing, this study adopts Hybrid Hierarchical Autonomous Learning Quantum Particle Swarm Optimization (HHQPSO) to optimize it. This paper studies the mathematical modeling of the production scheduling problem, defines the objective function and constraints, and then introduces the particle swarm optimization algorithm and improves it on this basis, and introduces the normal mutation operator. Then the population is dynamically divided by HHQPSO and different learning models are adopted to enhance the diversity of particles. An improved difference strategy is also proposed to update the random position of particles and Levy flight strategy is introduced. Finally, by comparing HHQPSO with traditional particle swarm optimization (PSO), this study found that the on-time delivery rate of HHQPSO was between 95.1% and 99.9%, which was significantly higher than that of PSO (89%–94.9%). In the aspect of inventory control, the warehouse occupancy rate of HHQPSO algorithm was 25% lower than that of PSO algorithm, which showed better inventory management ability. The results show that HHQPSO algorithm can improve production efficiency and reduce costs, and it plays an outstanding role in production scheduling optimization.