Abstract Manager

There are three quality levels (Levels 1 to 3) of asphalt mix inputs for designing and analyzing flexible pavements using the AASHTOWare Pavement ME Design (PMED) software. For the most reliable outcomes, Level 1 inputs (e.g., asphalt binder complex modulus and phase angle, mix dynamic modulus, indirect tensile strength and creep compliance) are recommended, which can only be obtained through comprehensive laboratory tests. Level 2 inputs can be derived from a limited number of laboratory tests. Level 3 inputs consist of typical physical properties (e.g., aggregate gradation, asphalt mix unit weight, air voids and binder content and asphalt binder grade), which are considered to provide the least reliable results. However, costly and sophisticated equipment, highly skilled personnel and significant time are required to obtain Level 1 input data, which are challenging for the PMED software users. Consequently, Level 3 inputs are generally used.

To compare and better understand the impacts of Level 1 and Level 3 inputs, the TAC ME Design Subcommittee conducted a series of design trials using inputs from five asphalt mixes containing 0 to 40% reclaimed asphalt and 11 weather stations across Canada. Preliminary analyses of these trial results indicated that Level 1 inputs predict lower IRI values in warmer climates and higher IRI values in colder climates than Level 3 inputs. Level 1 inputs had no impact on the predicted thermal cracking in colder climates but showed some impacts in warmer climates. In contrast, Level 3 inputs resulted in some variation in thermal cracking in colder climates while having no impact in warmer climates. Level 1 inputs resulted in lower AC and total rutting  compared to Level 3 inputs in both colder and warmer climates. Both Level 1 and Level 3 inputs predicted higher AC and total rutting in colder climates and lower rutting in warmer climates, which are unexpected. The predicted bottom-up fatigue cracks were higher and more sensitive when using the Level 1 inputs as compared to Level 3 inputs. PMED software predicted fewer top-down fatigue cracks in warmer climates, with greater sensitivity to change in input levels (Level 1 vs Level 3), than in colder climates.

The objective of this paper is to present the results of the above-mentioned trials, along with detailed  analyses and findings. The information provided may assist agencies and interested individuals in assessing the suitability of the PMED software and the impact of input quality on flexible pavement design.