Before joining the U of A faculty in 1996, he served in the US Army as an engineer officer for 24 years. During his military career Dennis had the unique opportunity to build roads, airfields and other facilities on five different continents and spend over 11 years as a member of the faculty at the US Military Academy. His current research interests include laboratory and field determination of geotechnical material properties for transportation systems and the use of remote sensing techniques to categorize geohazards. He has published over 85 peer reviewed articles relating to his research and educational activities. Dennis holds BS and MS degrees in Civil Engineering from the

No abstract

No abstract

Currently, most low-volume roads (LVRs) are designed on the basis of empirical experience or a loose implementation of some design procedure such as that found in the AASHTO guide. Procedures such as that of AASHTO were developed primarily for higher-volume roads and may not provide cost-effective designs for LVRs. A complicating factor in the AASHTO procedure is that it defines resilient modulus, an elastic material property that is expensive to measure, as the primary material property of interest for the design of pavements. This property is used despite the fact that many LVR pavement systems are subjected to nonelastic deformations. There is, therefore, a strong motivation to develop a sound design procedure applicable specifically to LVRs that uses simple soil properties. In this study, correlations to plastic deformation were made by using simple soil index properties. These correlations were based on measurements from more than 500 specimens of 20 Arkansas subgrade soils. Several finite element analyses were conducted to establish stress states in a pavement system from a variety of geometries and vehicle loads, and the information was tabulated. With this design procedure, a designer simply has to enter the design vehicle and some simple index properties to obtain rutting as a function of load repetitions. On the basis of this information, the user could specify a pavement configuration to carry a certain repetition of the design load or, given a pavement configuration, the model could evaluate damage from overloads or establish maximum loads during periods of poor sub-grade performance.