Three different computer analyses of the aasho road test for rigid pavements are reviewed, along with their resultant pavement performance equations. A fourth analysis, which is a revision of one of the three, is then briefly presented. A review is provided of the original aasho least-squares regression analysis of the serviceability data collected during the 2 years of the road test. In their book, "road work", small and winston propose that the original aasho analysis overestimates the life of thick rigid pavements and that a right-censored survival regression analysis would be more correct. Most ofthe thicker rigid test sections of the aasho road test were incorporated into interstate 80 and received 12 additional years of heavy-truck traffic. A review is provided of a report by the illinois department of transportation (dot) in which little and mckenzie reanalyzethe aasho data, including the additional traffic, using the least-squares method of regression analysis. For comparison, the survival regression analysis proposed by small and winston is reanalyzed usingthe additional data collected by illinois dot on the rehabilitated aasho roadway. It was concluded that the small/winston analysis significantly underestimated the life of thick rigid pavements. The original aasho method overestimated the life of thick rigid pavements, but not as significantly as the small/winston model underestimated it. A proposed revision to the small/winston model results in a performance prediction for thick rigid pavements slightly greater than thelittle and mckenzie revision to the original aasho analysis. Because of the lack of distress for the thicker rigid pavements in the road test, the small/winston survival analysis for the aasho road test rigid test sections is not valid. This paper appears in transportation research record no. 1307, Pavement analysis, design, rehabilitation, and environmental factors 1991.
Abstract