Multioutput Regression for Reinforced Retaining Wall Optimum Design with Machine Learning

AYDIN Y., BEKDAŞ G., NİGDELİ S. M., Işıkdağ Ü.

Studies in Systems, Decision and Control, Springer International Publishing Ag, ss.93-102, 2026 identifier

  • Yayın Türü: Kitapta Bölüm / Araştırma Kitabı
  • Basım Tarihi: 2026
  • Doi Numarası: 10.1007/978-3-032-07738-7_5
  • Yayınevi: Springer International Publishing Ag
  • Sayfa Sayıları: ss.93-102
  • Anahtar Kelimeler: Decision tree, Machine learning, Multioutput regression, Optimization, Retaining wall
  • İstanbul Üniversitesi-Cerrahpaşa Adresli: Evet

Özet

This study aims to predict optimum reinforced concrete wall dimensions and cost by using the height of the wall (H) and surcharge load (q) inputs using Multioutput Regression. The base learner models for Multioutput Regression used in this study are Decision Tree Regression (DTR), Support Vector Regression (SVR), Elastic Net Regression (ENT) and Histogram Gradient Boosting Regression (HGBR). Coefficient of Determination (R2), Mean Absolute Error (MAE) and Mean Squared Error (MSE) are used for performance evaluation. Among these models, DTR showed the highest prediction performance (R2: 0.855, MSE: 3930.22, MAE: 18.09).