An integrated scheduling model for single-machine systems considering fatigue dynamics, speed control, maintenance, and learning-based performance

• Proposes a single-machine scheduling model combining fatigue dynamics, learning, preventive maintenance, minimal repair, and speed-dependent emissions. • Models operator fatigue and learning over time, while enforcing a CO 2 emission cap linked to machine speed. • Includes proactive, preventive, and reactive minimal repair strategies with Weibull-based machine aging. • A tailored GA efficiently solves the model, utilizing fatigue-aware decoding and local improvement. • Numerical experiments show that integrating rest, maintenance, and learning significantly improves the makespan, reduces workload, and maintains emissions within limits.