Geotechnical specification/geology correlation model of soil environmental patterns: Evaluation metrics and NS, AIC/BIC criteria
• CEBs from soils of microlithic rocks are less dense than that granular one. • NS shows the adequacy of the models CBR - rock patterns origin. • The reduction in porosity in basalt clays is linked to the absence of quartz grains. • The firing of basalt soils produces more resistant crystals. • The values of MAE and MSE tend towards 0, the models are fair and accurate. Studies of earth construction materials take lengthy periods due to a lack of understanding of the relationships between geological and geotechnical characteristics. With a focus on the physico-mechanical, mineralogical, and geochemical characteristics of earth materials, this study aimed to propose specifications and a correlation model between geotechnical parameters, the geological origin of the soil, and its use for Compressed Earth Bricks (CEB) and pavement layers. Geotechnical parameters determined were identification tests for soils, shrinkage, absorption, compression, bending, as well as geochemical and mineralogical analyses. Four correlation patterns were defined according to petrographic origin, geological history, mineral associations, climate, and topography. The polynomial regression plots the compressive strength (MPa), flexural strength (MPa), absorption (g/cm 2 .min 0.5 ), and Californian Bearing Ratio (CBR) curves as a function of the percentages of addition of cement, lime, and firing temperature. Evaluation criteria, based on their parameters, were developed for 36 CEB formulation models and 12 pavement layer formulation models. Additionally, the NS and AIC/BIC criteria were determined. These tend toward 0, indicating that the models are reliable and precise. The relationships are similar to those that follow. For lateritic clay soils developed on trachybasalt (Pattern I), treated with cement for CEBs (x: percentage of cement, y: compressive strength of CEB in MPa), we have, y = 0.0357x 2 + 0.1877x + 1.3; with R² = 0.9191, MAE = 0.152, MSE = 0.029, RSME = 0.170, RMSLE = 0.026, NS = 0.919. For lateritic clay soils developed on granitoid (Pattern II), treated with lime for CEBs (x: percentage of lime, y: Flexural strength of CEB in MPa, we have, y = 0.0043x 2 + 0.4523x – 0.098; with R² = 0,948; MAE = 0.134, MSE = 0.025, RSME = 0.157, RMSLE = 0.034, NS = 0.949. For lateritic clay soils developed on orthogneiss (Pattern III), treated at high temperature for CEBs (x: temperature in°C, y: Absorption of CEB in g/cm 2 /min. 0.5 ), we have, y = -0.0335x 2 + 0.1165x + 1.254; with R 2 =1, MAE=0, MSE=0, RSME=0, RMSLE=0, NS=1. For lateritic clay soils developed on anatexite (Pattern IV) with 4 days of immersion in water, treated with cement for pavement layers (x= percentage of cement, y= CBR): y = -4.75x 2 + 30.05x – 8.25; R² = 0,9412, MAE = 1.9, MSE = 4.512, RSME = 2.124, RMSLE = 0.027, NS = 0.941. The AIC/BIC criteria range from -32.91 to 50.98.
Institutions
- Université de Dschang (CM)
- University of Namibia (NA)
- Université Officielle de Bukavu (CD)
Publication Details
- Journal
- Results in Engineering
- Published
- 2025-09-25
- DOI
- https://doi.org/10.1016/j.rineng.2025.107450
- Citations
- 1
- Primary Topic
- Agriculture, Water, and Health
- Type
- article
- Field-Weighted Citation Impact
- 0.82