Comparison of IDEAL‐CT Asphalt Cracking Tolerance Index Derived From Marshall and Superpave Gyratory Compacted Specimens

National Cooperative Highway Research Program (NCHRP) Project 20‐07 evaluated the Indirect Tensile Asphalt Cracking Test (IDEAL‐CT) as practical, reliable, and efficient for mixture design and production quality control, contributing to its rapidly increasing adoption. IDEAL‐CT utilizes monotonic loading of cylindrical specimens at intermediate temperatures to characterize cracking resistance based on a fracture mechanics parameter known as cracking tolerance index (CT _Index ). The test standard, ASTM D8225‐19, specifies Superpave gyratory compacted specimens. However, the CT _Index equation has specimen dimensions as input variables to allow testing of other specimen sizes. While the influence of mixture characteristics on CT _Index has been studied, the impact of specimen size remains unexplored, although it can affect strength properties. To enhance the utility of IDEAL‐CT, this study analyzed CT _Index derived from Marshall‐ and SGC‐compacted test specimens, providing insights to guide specimen fabrication, defining test criteria, and converting between CT _Index obtained from the two specimen types. Ten plant‐produced mixtures were subjected to IDEAL‐CT testing to provide 24 CT _Index values. On average, the Marshall‐based CT _Index exceeded the SGC‐based CT _Index by 42%, with greater variability, influenced by the differences in compaction mechanism and specimen size. A weak linear relationship between the two CT _Index groups was observed, suggesting that IDEAL‐CT inadequately addressed the impact of specimen size variations, despite including specimen dimensions in the CT _Index equation. Statistical testing at α  = 0.05 suggested the mean CT _Index difference was insignificant. However, effect size analysis using Cohen’s d and the Pearson correlation coefficient indicated the mean CT _Index difference was practically significant. Marshall‐based CT _Index should not be interchanged with SGC‐based CT _Index without mixture‐specific validation, as potential large differences in CT _Index could result in erroneous conclusions.