A Comparative Study on IW and Annealing-Based Coordinate Prediction Methods Using Small-Scale Geotechnical Property Data

Hyung-Koo Yoon

doi:10.7843/kgs.2025.41.6.119

All Issue

2025 Vol.41, Issue 6 Preview Page Next Page

A Comparative Study on IW and Annealing-Based Coordinate Prediction Methods Using Small-Scale Geotechnical Property Data 소규모 지반 물성 자료를 활용한 IW 및 어닐링 기반 좌표 예측 기법의 비교 연구

31 December 2025. pp. 119-129

PDF XML

Abstract

The kriging method, which is widely used to estimate data at unmeasured locations in geotechnical investigations, exhibits a notable limitation because its reliability significantly decreases as the spatial correlation between observation points diminishes with distance. The objective of this study was to enhance the spatial prediction accuracy of soil depth within a small-scale geotechnical dataset by performing coordinate prediction based on Independent Weighted (IW) and Annealing algorithms. A total of 23 soil depth measurements were collected through the Dynamic Cone Penetration Test, and data sparsity was addressed using the synthetic minority over-sampling technique and adaptive synthetic sampling. The IW algorithm calculates soil depth at each grid point using a distance-weighted average for the initial prediction, while the Annealing algorithm takes the IW results as an initial solution and iteratively performs probabilistic swap operations to minimize overall error. Comparative analysis shows that after applying Annealing, the prediction error (RMSE) decreased by approximately 25–30%, and the normalized error sum (Energy) decreased by more than 40% compared to IW. Furthermore, six performance evaluation metrics were applied, all demonstrating improved reliability following the application of Annealing. These findings confirm that the proposed IW and Annealing methods provide an efficient and reliable approach for realistically reproducing soil depth distributions, even in small-scale geotechnical investigation environments.

Keywords

Annealing

Coordinate

Independent Weighted

Soil depth

지반의 미측정 지점의 데이터를 유추할 수 있는 크리깅 기법은 관측 지점 간 거리가 멀어질수록 상관성이 급격히 감소하여 신뢰성이 낮은 한계를 보인다. 본 연구의 목적은 소규모 지반조사 데이터인 토심을 이용하여 공간적 예측 정확도를 향상시키기 위한 Independent Weighted(IW) 및 어닐링(Annealing) 기반으로 좌표 예측을 수행하는 것이다. Dynamic Cone Penetration Test(DCPT)를 통해 총 23개 지점에서 실측된 토심 데이터를 활용하였으며, 데이터의 희소성을 보완하기 위해 Synthetic Minority Over-sampling Technique(SMOTE)와 Adaptive Synthetic Sampling(ADASYN) 알고리즘을 적용하였다. IW는 각 격자점의 토심을 거리 가중 평균으로 계산하여 초기 예측을 수행하였고, 어닐링은 IW 결과를 초기 해로 설정한 후 확률적 스왑 연산을 반복 수행하여 전체 오차를 최소화하였다. 비교 결과, 어닐링 적용 후 예측 오차(RMSE)는 IW 대비 약 25~30% 감소, 정규화된 오차합(Energy)은 약 40% 이상 감소하였다. 또한 성능을 평가할 수 있는 6가지 지표 값도 제시하였으며, 모두 신뢰성이 향상된 결과를 보여줬다. 해당 연구에서 사용한 방법인 IW 및 Annealing 기법은 소규모 지반조사 환경에서도 토심 분포를 현실적으로 재현할 수 있는 효율적이고 신뢰성 높은 예측 방법임이 검증되었다.

References

Christianson, R. B., Pollyea, R. M., and Gramacy, R. B. (2023), “Traditional Kriging Versus Modern Gaussian Processes for Large-scale Mining Data”, Statistical Analysis and Data Mining: The ASA Data Science Journal, 16(5), 488-506, https://doi.org/10.1002/sam.11635.

10.1002/sam.11635

Ghannadi, P., Kourehli, S. S., and Mirjalili, S. (2023), “A Review of the Application of the Simulated Annealing Algorithm in Structural Health Monitoring (1995-2021)”, Fracture and Structural Integrity, 17(64), 51-76.

10.3221/IGF-ESIS.64.04

Hong, S., Lee, M., Ko, S. J., Kwak, T. Y., and Kim, S. R. (2024), “Analysis of Multiple Correlations Between Ground Information and Final Settlement of Soft Ground Using Ensemble-Based Models”, Journal of the Korean Geotechnical Society, 40(6), 111-124.

10.7843/KGS.2024.40.6.111

Ji, Y., Kim, H. S., Lee, M. G., Cho, H. I., and Sun, C. G. (2021), “MLP-based 3D Geotechnical Layer Mapping Using Borehole Database in Seoul, South Korea”, Journal of the Korean Geotechnical Society, 37(5), 47-63.

10.7843/KGS.2021.37.5.47

Kim, S. Y., Lee, J. S., and Hong, W. T. (2021), “Subgrade Assessment Using Automated Dynamic Cone Penetrometer to Manage Geo-infrastructures”, Smart Structures and Systems, 27(5), 861-870.

10.12989/SSS.2021.27.5.861

Kim, S. Y., Lee, J. S., Park, G., and Hong, W. T. (2022), “Evaluation of Dynamic Resistance for Application of Portable in-situ Device to Extra Target Depth”, KSCE Journal of Civil Engineering, 26(10), 4195-4201.

10.1007/s12205-022-2031-z

Liu, Z. N., Yu, X. Y., Jia, L. F., Wang, Y. S., Song, Y. C., and Meng, H. D. (2021), “The Influence of Distance Weight on the Inverse Distance Weighted Method for Ore-grade Estimation”, Scientific Reports, 11(1), 2689, https://doi.org/10.1038/s41598-021-82227-y.

10.1038/s41598-021-82227-y33514843PMC7846575

Meng, D., Yang, H., Yang, S., Zhang, Y., De Jesus, A. M., Correia, J., and Zhu, S. P. (2024), “Kriging-assisted Hybrid Reliability Design and Optimization of Offshore Wind Turbine Support Structure based on a Portfolio Allocation Strategy”, Ocean Engineering, 295, 116842, https://doi.org/10.1016/j.oceaneng.2024.116842.

10.1016/j.oceaneng.2024.116842

Pang, Y., Wang, Y., Lai, X., Zhang, S., Liang, P., and Song, X. (2023), “Enhanced Kriging Leave-one-out Cross-validation in Improving Model Estimation and Optimization”, Computer Methods in Applied Mechanics and Engineering, 414, 116194, https://doi.org/10.1016/j.cma.2023.116194.

10.1016/j.cma.2023.116194

Venkateswaran, C., Ramachandran, M., Ramu, K., Prasanth, V., and Mathivanan, G. (2022), “Application of Simulated Annealing in Various Field”, Materials and its Characterization, 1(1), 01-08.

10.46632/mc/1/1/1

Xiao, M., Huang, S., and Chen, X. (2024), “Maximum Weighted Independent Set: Effective Reductions and Fast Algorithms on Sparse Graphs”, Algorithmica, 86(5), 1293-1334, https://doi.org/10.1007/s00453-023-01197-x.

10.1007/s00453-023-01197-x

Yang, M., Zhang, D., Wang, F., and Han, X. (2022), “Efficient Local Adaptive Kriging Approximation Method with Single-loop Strategy for Reliability-based Design Optimization”, Computer Methods in Applied Mechanics and Engineering, 390, 114462, https://doi.org/10.1016/j.cma.2021.114462.

10.1016/j.cma.2021.114462

Yoon (2023), “Method for Assessing Landslide Susceptibility Using SMOTE and Classification Algorithms”, Journal of the Korean Geotechnical Society, 39(6), 5-12.

10.7843/KGS.2023.39.6.5

Information

Publisher :The Korean Geotechnical Society
Publisher(Ko) :한국지반공학회
Journal Title :Journal of the Korean Geotechnical Society
Journal Title(Ko) :한국지반공학회 논문집
Volume : 41
No :6
Pages :119-129
Received Date : 2025-11-10
Revised Date : 2025-11-21
Accepted Date : 2025-11-22
DOI :https://doi.org/10.7843/kgs.2025.41.6.119

Journal of the Korean Geotechnical Society ISSN:1229-2427(Print) 2288-646X(Online) 한국지반공학회 논문집

All Issue