Interpreting SuperPAVE PG test results with confidence intervals

• Published in: Construction & building materials 2011-06, Vol.25 (6), p.2777-2784
• Main Authors: Aflaki, Sassan, Memarzadeh, Milad
• Format: Article
• Language: English
• Subjects: Bitumen selection; Bituminous materials; Confidence intervals; Iranian bitumen; Nonlinear regression; Performance grading; Properties; Regression analysis; Uncertainty
• ISSN: 0950-0618; 1879-0526
• DOI: 10.1016/j.conbuildmat.2010.12.059

► Nonlinear regression and confidence intervals are used for evaluation of bitumen PG. ► The formulation introduced in this paper could increase the reliability in bitumen selection. ► The binder’s fatigue and rutting parameters are evaluated by proposed formulation. The innovative method to interpret SuperPAVE PG test results using regression analysis and confidence intervals is evaluated in this research. Till now, most of researches and conclusions in the field of bitumen rheological aspects are based on the mean values of test results. Using the mean values does not show the reliability of the test results and in many cases two bitumen with the same PG show different performances, due to different data variances. In this study the confidence intervals method is implemented to interpret the results of SuperPAVE PG tests and notify that the mean values of the variables are occasionally misleading and always some uncertainty exists when using bitumen. This study presents a new formulation that quantifies the uncertainties in bitumen behavior to obtain the true temperature grade of the bitumen with different confidence levels (75% and 95%). Two samples of bitumen with same PG were evaluated with this formulation and the interesting results were found. The behavior of these samples with same PG was significantly different regarding the higher confidence intervals.