AbstractView references

Machining plays a crucial role in modern manufacturing, relying on automated processes to efficiently create complex parts through subtractive like lathe turning and cutting. However, a major concern in this manufacturing process is tool wear, necessitating a robust system for proactive malfunction detection. To keep up with advancements and meet the increasing demands of speed and precision, artificial intelligence (AI) emerges as a promising solution. However, AI algorithms often require fine-tuning of hyperparameters, which poses a challenge. Swarm intelligence algorithms, inspired by collaborative behaviors observed in nature, offer a potential solution. By applying swarm intelligence to hyperparameter optimization, AI algorithms can achieve optimized models that address time and hardware constraints. This work proposes a methodology based on Extreme Gradient Boosting (XGBoost) for forecasting malfunctions. Additionally, a modified optimization metaheuristic is introduced to specifically enhance the performance of this methodology. To evaluate the proposed approach, it has been applied to a real-world dataset and compared to several well-known optimizers. The results demonstrate admirable performance, highlighting the potential of swarm intelligence in achieving efficient and effective machining processes. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.