CD Tesis

Pemodelan Kuat Geser Tanah Lunak Pesisir Riau Berbasis Soft Computing

---

Coastal soils are dominated by marine clay and organic-rich soils with low shear strength, high compressibility, and sensitivity to moisture content and salinity, thereby increasing the risk of infrastructure failure. This study aims to identify soil parameters that influence undrained shear strength (Cu), analyze the relationships between parameters, and compare the performance of ANN, ANFIS, and empirical formulas in predicting Cu values. The data consist of 210 coastal soil samples, including moisture content, bulk density, specific gravity, plastic limit, liquid limit, plasticity index, and liquidity index. The ANN was implemented using a multilayer perceptron with a backpropagation algorithm, while the ANFIS combines neural networks and fuzzy logic to capture nonlinear relationships and data uncertainty. The results indicate that the parameters w, PL, LL, and PI have a strong influence on the Cu value. Correlation analysis reveals a strong relationship between soil properties and the soil shear strength, as demonstrated by the main equation: Cu = 0.337w + 0.236PL + 0.276LL + 0.201PI. The ANFIS method provides the best prediction accuracy (R = 0.99999; R² = 0.99997; MAPE = 0.146%) compared to ANN and empirical equations, making it more effective in predicting undrained soil shear strength (Cu).


Key words: soil physical properties, undrained shear strength, coastal areas, ANN, ANFIS

---

Ketersediaan

Informasi Detail

Judul Seri

-

No. Panggil

2410247172

Penerbit

Pekanbaru : Universitas Riau – Pascasarjana – Magister Teknik Sipil., 2026

Deskripsi Fisik

-

Bahasa

Indonesia

ISBN/ISSN

-

Klasifikasi

2410247172

Tipe Isi

-

Tipe Media

-

Tipe Pembawa

-

Edisi

-

Subjek

Magister Teknik Sipil

Info Detail Spesifik

-

Pernyataan Tanggungjawab

Vina

Versi lain/terkait

Lampiran Berkas

Komentar

---

Anda harus masuk sebelum memberikan komentar