Dimensioning of the Retaining Wall Using Linear Regression, Ridge Regression and Lasso Regresion

AYDIN Y., BEKDAŞ G., NİGDELİ S. M.

11th International Conference on New Technologies, Development and Applications, NT 2025, Sarajevo, Bosna-Hersek, 26 - 28 Haziran 2025, cilt.1484 LNNS, ss.483-490, (Tam Metin Bildiri) identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası: 1484 LNNS
  • Doi Numarası: 10.1007/978-3-031-95200-5_53
  • Basıldığı Şehir: Sarajevo
  • Basıldığı Ülke: Bosna-Hersek
  • Sayfa Sayıları: ss.483-490
  • Anahtar Kelimeler: cost, machine learning, optimization, prediction, retaining wall
  • İstanbul Üniversitesi-Cerrahpaşa Adresli: Evet

Özet

Retaining walls are engineering structures that hold the ground at two different levels and are under the influence of lateral pressures.Since the retaining wall is a common structure, its low-cost design is preferred.Parameters such as the geometric properties of the reinforced concrete retaining wall and the presence of surcharge load are effective in retaining wall design. Since retaining walls are designed to hold slopes, calculations are important. In this study, the optimum dimensions and cost of a retaining wall were predicted using the height of the wall (H) and surcharge load (q). For this purpose, Linear Regression (LR), Ridge Regression and Lasso Regresion are used as base learners for Multioutput Regression. The results are evaluated using performance metrics such as Mean Square Error (MSE), Mean Absolute Error (MAE), Coefficient of Determination (R2), Mean Absolute Percentage Error (MAPE).While Ridge Regresion (R2 = 0.79972) and Linear Regresion (R2 = 0.79979) performed similarly, Lasso Regresion performed the worst (foroptimum dimension and cost prediction of reinforced retaining wall.