A Bayesian framework for mechanical characterization of unstabilized rammed earth reinforced with polypropylene fibers
Abstract
• A mechanical characterization of fiber-reinforced rammed earth reinforced is presented • Compressive strength is shown to increase by 163.3 • A comprehensive probabilistic modelling framework is presented • Validation with an independent specimen confirmed accuracy of the framework Unstabilized rammed earth reinforced with fibers is being explored as an alternative to chemical stabilization, however, the mechanical response of these composites and the associated uncertainties remain insufficiently investigated particularly at higher fiber content. This study investigates the mechanical behavior of unstabilized rammed earth when reinforced with polypropylene fibers at 0%, 1.5%, 3.5%, and 5% of fiber content. To this end, an experimental campaign was carried out by testing unconfined compressive strength, three-point bending, and ultrasonic pulse velocity in a number of samples with various fiber percentages. Data show a clear trade-off: increasing fiber content substantially enhances compressive strength but reduces flexural strength and stiffness-related properties. Specifically, a fiber content of 5% increased the compressive strength by 163.3% (from 6.2 MPa to 16.4 MPa) while flexural strength decreased by 79.2%. In addition, phenomenological models of the various mechanical properties are inferred from the data using a rigorous Bayesian inverse problem framework. When simulated forward, these models enable the probabilistic estimation of compressive strength, elastic modulus, and material density using non-destructive ultrasonic pulse velocity measurements as input. This would enable in-situ property estimation and non-destructive quality control of polypropylene-fiber-reinforced rammed earth elements in practical engineering scenarios.
Fields of Study & Subjects
Hygrothermal properties of building materials
Geotechnical Engineering and Soil Stabilization
Masonry and Concrete Structural Analysis
Publication details
- Journal / source
- Results in Engineering
- Volume
- 29
- Pages
- 109085–109085
- Publication date
- 18 January 2026
- Language
- EN