Their current AI-based scheduling system considers factors like workforce availability, operation sequences, and machine setup times.

However, the system has limitations, particularly in handling complex scheduling requirements, and faces performance issues related to AI, prompting the need for improvement.

The Job Shop Scheduling Problem (JSSP) is a complex, NP-hard optimization challenge faced by industries where jobs need to be assigned to machines with sequence-dependent setup times, preemption, and other constraints. This research explores hybrid algorithmic approaches, including Tabu Search (TS), Adaptive Large Neighborhood Search (ALNS), Particle Swarm Optimization (PSO), and Ant Colony Optimization (ACO), to address JSSP in real-world production settings. By combining these methods with mixed integer programming, the studies aim to achieve near-optimal scheduling solutions that minimize factors like tardiness, and setup time. Results show that while these hybrid approaches perform well on small to medium datasets, their efficiency diminishes with larger datasets, indicating a need for further refinement to scale effectively for industrial applications.