Recent surveys have shown the two most common current processes used by city and county engineers to design asphalt pavements are the R-Value and Soil Factor procedures. The Soil Factor procedure is in the Mn/DOT State Aid guidebook, while the R-Value method is found in the Geotechnical Manual. These design methods are based on technology developed during the 1930s through early 1960s. The methods are based upon empirical performance data and material characterizations, which are a function of the available traffic and other conditions that existed 40 to 70 years ago.
Conditions have changed since then. For example, typical tire pressure at that time was on the order of 70 psi, whereas today it is on the order of 100 psi. Only conventional asphalt concrete surfaces were used then, but modifiers are now being used increasingly in mixtures. Therefore, more rational approaches to thickness design are needed in order to readily accommodate future changes in traffic conditions and material types. Such approaches would have to be based on the physical response of the pavement (i.e., stress, strain, or displacement) as opposed to an empirical relationship.
In the last few years, researchers at Mn/DOT, in conjunction with the University of Minnesota, have developed a mechanistic-empirical (M-E) design program called MnPAVE for flexible pavements using data from the Mn/ROAD research facility.
"The concept of M-E design has been discussed by engineers for 30 to 40 years," said Dave Van Deusen, pavement design engineer at Mn/DOT. The difficult issue with pavement design is that there are so many variables to consider, from the traffic volume on the road to the material properties of the pavement layers. Because there are so many variables, to create purely empirical formulas that will work under a variety of circumstances would be virtually impossible. With the advent of MnPAVE, which is based on the linear elasticity for a layered system, a designer can quickly and accurately analyze the behavior of layered pavements.
Mn/PAVE is a computer program that combines known empirical relationships with a more accurate representation of the physics and mechanics behind flexible pavement behavior. Research from Mn/ROAD has showed that flexible pavement structures can be modeled as linear elastic, multilayered systems. With this in mind, researchers chose to examine two distress types: rutting and fatigue cracking. Relationships were found between the onset of these two distresses and the number and type of load repetitions. These relationships are expressed in the form of transfer functions, which relate pavement structural response to distress and form the empirical portion of the process. One of the difficulties often encountered when dealing with transfer functions is the inability to obtain functions in a laboratory environment consistent with those in the field. Mn/ROAD is especially valuable in validating transfer functions.
"The mechanistic portions of the program rely on calculating the tensile strains at the bottom of the asphalt layer and the compressive stress at the top of the subgrade," explained Dave Timm, a University of Minnesota researcher partially responsible for the creation of ROADENT, MnPAVE's predecessor.
Research for this portion of the program was conducted almost exclusively using Mn/ROAD data. Projects done by former U of M professor Dave Newcomb and Bruce Chadbourn, a project engineer at Mn/DOT, where measured strains were compared with those predicted by the structural model, were highly influential.
The format of MnPAVE consists of three design levels: basic, intermediate, and advanced. Which of the three different levels used depends on the amount and quality of information known about the material properties and traffic data. The intermediate level corresponds to the amount of data currently required for Mn/DOT projects. The newer, and most useful, portions of the program lie in the other two categories. In the basic mode, only a general knowledge of the materials and traffic data is required for a reasonable yet conservative pavement design to be generated. MnPAVE's most efficient mode will be the advanced category. This design level requires the determination of modulus values for all materials over the expected operating range of moisture and temperature. Because of the dramatic changes that can occur in modulus values from the end of winter to the beginning of summer, spring has been divided into beginning and late periods. Pavements designed using the advanced feature should have a very predictable life span.
Although the MnPAVE program is still being validated and calibrated, it is quickly becoming clear that it will offer many useful features not previously accessible to road pavement designers. The M-E approach will allow engineers to quantify the benefits of using improved materials and specifications. For example, MnPAVE could demonstrate the effect of substandard compaction techniques, said Timm. Other benefits include the ability to address varying stiffnesses of performance-graded asphalt binders, the use of recycled materials, and increasing truck weights. This kind of information is extremely valuable for improved performance prediction and can lead to a more accurate lifecycle cost analysis. In addition, the software will be very versatile. Traffic data can be quantified using load spectra or ESALs, and the user can choose English or SI units.
There is some concern that a new pavement design technique will radically alter the current designs. However, researchers have offered assurances that this will not be the case. In fact, the most probable outcome is that most designs can be more accurately applied to specific situations, will be more efficient, and will require slightly less material. An additional related benefit to M-E design is that the program can account for the effects that seasonal temperature and variations have on material properties. Previous design techniques were especially lacking in this area, in that they were only able to consider either an average or worst case temperature and moisture.
Research and development of new pavement design procedures for both flexible and rigid pavement is being done under NCHRP Project 1-37A, Development of the 2002 Guide for the Design of New and Rehabilitated Pavement Structures. Upon completion, the 2002 guide is anticipated to become the latest, and most advanced, revision of the AASHTO Design Guide.
Project 1-37A is utilizing materials and performance data from Mn/ROAD sections in addition to other test facilities from around the nation. "The structural model in both programs should be the same, but the performance model used by AASHTO may be different," said Dr. Shongtao Dai, a research operations engineer at Mn/DOT. When the 2002 guide is released, the DOT will evaluate it with an eye toward possible adoption of the procedure. Another possibility would be to alter MnPAVE based on the recommendations of NCHRP 1-37A. In either case, the work done in developing and implementing MnPAVE will provide Minnesota with a head start on the M-E design process and make the transition process much more effective.
Nate Erpestad
(For more information on MnPAVE, please contact Mn/DOT information at 651-296-3000 or Nate Erpestad with the Center for Transportation Studies at 612-626-2862.)