Variability and design partial factors for rubberized concrete based on experiments and a large-scale database

In recent years, the use of waste rubber as a concrete aggregate has attracted increasing attention in the context of carbon neutrality. The mechanical performance of rubberized concrete (RuC) is strongly influenced by the amount and characteristics of the incorporated rubber; however, the variability of compressive strength and its implications for structural design have not yet been systematically clarified. In this study, a comprehensive database comprising 183 test results was constructed by integrating newly obtained experimental data with previously published results. Using this database, the effects of rubber properties and mixing conditions on the characteristics of both fresh and hardened RuC were systematically analyzed. The results indicate that fresh concrete properties, such as slump, exhibit only weak correlations with rubber content and particle size, suggesting that workability cannot be reliably predicted from rubber quantity alone. In contrast, the variability of compressive strength increases systematically with increasing rubber volume replacement ratio. The coefficient of variation of compressive strength was explicitly formulated as a function of the rubber volume ratio, enabling a quantitative description of strength variability induced by rubber incorporation. Using this formulation, reliability-based partial factors dependent on the rubber volume ratio were derived. The proposed approach allows the compressive strength of RuC with different rubber contents to be designed to a consistent target safety level, providing a practical basis for reliability-oriented structural design of RuC. • Effects of rubber content and properties on fresh and hardened concrete evaluated. • Compressive strength variability is quantified as a function of rubber volume ratio. • Reliability-based partial factors accounting for strength variability are proposed. • Design-oriented strength model ensuring target failure probability is established. • Integrated framework linking mix design, strength prediction, and reliability.