AI-Assisted Optimization Approach Based on XGBoost+NSGA-III+TOPSIS for Milling of Plastic Mold Steel
2026 International Conference on Electrical, Computer, and Energy Technologies, ICECET 2026, Rome, İtalya, 6 - 09 Temmuz 2026, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/icecet65726.2026.11632560
- Basıldığı Şehir: Rome
- Basıldığı Ülke: İtalya
- Anahtar Kelimeler: Decision making, Optimization, Plastic mold steel, Prediction
- Yıldız Teknik Üniversitesi Adresli: Evet
Özet
Milling of plastic mold steels involves strongly conflicting objectives, such as maximizing tool life, surface quality, and material removal rate simultaneously, which cannot be satisfactorily addressed by conventional parameter selection methods due to their nonlinear and trade-off-driven interactions. In this study, a data-driven multi-objective optimization framework was established to determine the optimal cutting parameters, regarding surface roughness, flank wear, and material removal rate, for plastic mold steel milling operation. The framework combines Extreme Gradient Boosting (XGBoost) as a surrogate model with the Non-dominated Sorting Genetic Algorithm III (NSGA-III) as the optimizer and Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) as the decision maker. The model predicted tool wear, surface roughness, and material removal rate outputs at percentages of 81.74%, 85.62%, and 97.46%, respectively, and the parameter set that optimized all outputs was determined to be a cutting depth of 0.24 mm, a feed value of 0.13 mm/rev, and a cutting speed of 154.1 m/min.