Researchers have developed a novel surrogate-assisted genetic programming method to enable lifelong learning for dynamic flexible job shop scheduling (DFJSS). This approach aims to overcome the limitations of traditional genetic programming, which often suffers from catastrophic forgetting when adapting to new tasks or changing job characteristics. The proposed framework balances plasticity and stability, regulates knowledge reuse based on task similarity, and uses an improved surrogate model with distance constraints for better accuracy. Experiments demonstrate that this method effectively reduces knowledge loss while maintaining adaptability, outperforming existing techniques in dynamic scheduling scenarios. AI
IMPACT This research could lead to more adaptable and robust scheduling systems in dynamic environments, reducing performance degradation and knowledge loss.
RANK_REASON Academic paper detailing a new method for lifelong learning in genetic programming for scheduling. [lever_c_demoted from research: ic=1 ai=1.0]
- arXiv
- Catastrophic interference
- Dynamic Flexible Job Shop Scheduling
- genetic programming
- Knowledge reuse: temporal-abstraction mechanisms for the assessment of children's growth.
- lifelong learning
- plasticity
- surrogate model
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
