Statistical Analysis of Influence of Mix Design Parameters on Mechanical Properties of Mixes with Reclaimed Asphalt Pavement

Abstract

The objective of this study was to evaluate statistically whether and how various mix design parameters influenced the complex modulus and fatigue resistance of asphalt mixes produced with reclaimed asphalt pavement material. Sixteen asphalt mixes were tested by following a 25-1 fractional factorial experimental design. Five mix design variables (considered as two-level factors) were selected: aggregate nature (limestone versus basalt), grading curve (continuous versus gap-graded), filler nature (limestone versus hydrated lime), binder content (4.35% versus 5.35%), and binder nature (35/50 versus 35/50 B). Analysis of variance was performed on experimental results obtained for three responses (used in the French pavement design guide): complex modulus |E*| (158C, 10 Hz) and fatigue parameters ∊6 and 1/b (10°C, 25 Hz). Main effects of all factors and their interactions with binder content were monitored. All factors appeared to have had a statistically significant influence on complex modulus |E*|. All factors except filler nature showed a significant effect on fatigue parameter ∊6. For both complex modulus |E*| and fatigue parameter ∊6, a significant interaction was observed between the effects of binder content and binder nature. For this reason, the influence of those two factors was mutually dependent, and they could not be evaluated separately. No conclusions could be drawn about the influence of mix design factors on fatigue parameter 1/b.