Soil Consistency Prediction Based on Cone Penetration Test (CPT) Using ANN (Artificial Neural Network) and Multinomial Regression (Case Study: Surabaya Region)

• Published in: Lecture Notes in Civil Engineering
• Main Author: Wahyuni F.; Wildani Z.; Purnamasari R.
• Format: Conference paper
• Language: English
• Published: Springer Science and Business Media Deutschland GmbH 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85200354705&doi=10.1007%2f978-981-97-0751-5_51&partnerID=40&md5=4bf4558afb623113b422fcce1c8cc40a
• ISSN: 23662557
• DOI: 10.1007/978-981-97-0751-5_51

The development of Surabaya city can be seen from the many developments in the town. Many soil investigation tests have been carried out with the many products that have been and will be carried out in Surabaya. One of the soil investigation tests is the Sondir test, the CPT (cone penetration test). CPT is a soil investigation method that is quite affordable and easy to do, which produces parameters in the form of soil consistency at each soil depth. Based on soil data from 2017 to 2021, CPT data is spread across various areas in Surabaya. Thus, this research aims to find predictions of soil consistency in the Surabaya area from existing soil data using the ANN (artificial neural network) and multinomial logistic regression methods. From this study, it was found that from 3196 scattered soil data, it showed that the coefficient of determination (R2) was 98.44%. This means that the proportion of the variability of Depth (m), Qonus value (qu), and Area can explain the consistency of the soil and is an excellent model. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.