Comparing Design Strain to Actual Strain on New Roads

Abstract

In an active design project it was found that the target design strain was fulfilled for about 90 % of the several hundred tests made with a Falling Weight Deflectometer (FWD). The tests were done after one year of traffic. This outcome is in line with current design principles employing statistical methods based on reliability. More disturbing is the fact that about half of the sections would have an expected life that exceeds twice the design life. This was in spite the active design efforts to reduce bearing capacity variability. From a structural point of view, this may not be alarming; but from a sustainability standpoint one could see resources being wasted on sections that do not need such a well-built design. Thus, it would be of interest to see what the bearing capacity is on new road sections and compare it with the design code. In 2010 the Swedish Road Administration funded a project where a number of new and widened roads were tested for bearing capacity with a FWD. The roads comprised different categories and types. The tests showed that generally the roads are overdesigned as compared to the design code. The present paper discusses whether this is purely an effect of calibration in the mechanistic-empirical process or if there are any inherent errors in the code handling the natural variability. If the latter is the case, investment costs could be lowered for the benefit of sustainability. Further, more efforts could be placed on the sections exceeding the design strains.