Use of PMS Data for Performance Monitoring with Superpave as an Example
11. Conclusions, Findings and Recommendations
This section summarizes project findings, conclusions, and recommendations. The project was initiated with a team balanced in its understanding and experience in pavement management, the Superpave Design System, and interaction with AASHTO and state DOTs. An early meeting was held with the sponsors to discuss the various aspects of the project. As a result of this meeting a work plan was laid out and the states of Maryland, Indiana, Florida, Arizona and Washington were selected for visits. These states were selected because of their experience with pavement management and with Superpave.
The Project Team reported their findings to the FHWA, and as a result of this the project was extended with a Pathfinder Study in which Maryland SHA was asked to participate. This study was intended to serve as an example of how a DOT could identify and collect required data on Superpave and how much effort would be needed to enter these into one or more electronic database. As a next step these databases were loaded into a suitable "database" for storage, inspection, linking, analysis and reporting purposes. In this case, use was made successfully of a recently developed website of the University of Washington.
As a result of these visits, activities, analysis and interaction among the staff and sponsors, the following conclusions have been derived.
- These project activities show that it is possible for state DOTs to assemble a database or a data warehouse that can be used to evaluate Superpave performance, other design, and new materials concepts. The emphasis in this project was not to completely evaluate Superpave, but to determine the feasibility of the concept and its applicability among several states.
- Washington State DOT has, in collaboration with The University of Washington, developed an approach whereby relevant data for Superpave contracts are made available on a website. The performance, design and construction data can be organized, downloaded and analyzed. This new development was possible because nearly all their data on materials and construction are now available in electronic format (mostly Excel files), and in addition a major effort has been made to link these data to performance measurements from their PMS. The project is still in a test phase, but it looks promising, because it could be easily extended to other materials and techniques. For this project, and for the Pathfinder Study in particular, successful use was made of this new methodology, with the help of data supplied by Maryland SHA.
- A major advantage of the website approach is that all data are available to all users as soon as they are entered. When the proper equipment for electronic data entry is used in the field, it is possible to monitor construction projects instantly with "real time" QC/QA data which could be very beneficial.
- The Pathfinder study has shown that the collection of relevant data in materials and construction files, required for linking with performance data, can be cumbersome and time consuming. In Maryland it took two man-months to collect data for 7 Superpave projects, and even then not all required data could be found and entered.
- All relevant data should ideally be put into electronic format from the start, but there is also a need for a proper and unbiased definition of those data before they can be used in a linking exercise.
- The many details and variables involved in a new broad-based approach such as Superpave requires recording detailed data on design, construction, and performance data for several individual projects for comparison. These details are beyond normal pavement management data activities, but can be added as a supplement and the results are well worth the effort. As the data are extended over five - ten years, analysis results and updates to the performance models can be substantial and the results highly beneficial.
- The key to linking databases for performance, materials and construction is to have precise common location identification and date/time information. Only in this way can it be assured that the data are comparable. Unambiguous locations can be provided by GPS measurements and these are relatively economical at the present time. However, they must be tied to traditional location identification information such as project number, mile point, lane, direction, date, etc.
- The best approach to using pavement management and related data to evaluate new design and material concepts such as Superpave, will involve several states setting up databases with the required detailed data and combining their efforts to make the necessary performance evaluations with a team of 5-25 states. This will require coordination among states, with possible FHWA and/or AASHTO support.
- This review of information data, and support from these five state DOTs, suggests that an extended effort among several states in the field of evaluation of new concepts such as Superpave is warranted. In the specific case of Superpave more than $100 million has been expended to date and the comparison of actual to projected performance using the new specifications and methods essential to state DOTs to prove the concepts.
- Preparation of a good conceptual work plan for evaluating new concepts using PMS data and such tools as standard Pavement Evaluation Protocols will be useful information for state DOTs to encourage them to set-up appropriate evaluation databases and procedures for any new pavement concept they undertake.
- The approach examined in this project for evaluation Superpave is warranted for other new concepts and a project should be set up to outline a methodology for state DOT use.
The results of this project and the interaction with the five state DOTs supports the following findings:
- All state DOTs contacted during this project showed a strong interest in using their pavement management data combined with materials and construction data to evaluate new concepts such as Superpave.
- All five state DOTs visited in this project had adequate PMS data that can be used for a proposed Superpave evaluation.
- At the present time, most of the states visited don't have an appropriate electronic format for other required materials and construction data needed for the evaluation analysis. However, these states have the data available in field files, which could be converted to electronic format with appropriate time and manpower effort. Most data required for proper analysis are available with the notable exception of the actually applied layer thickness.
- Of the five state DOTs visited, only WSDOT and MDSHA currently have direct linkages between their materials & construction data and their PMS database for a number of Superpave projects. In every case the available personnel who could link these databases were submerged in their other duties. In some cases personnel cutbacks have limited the state's ability to do the necessary work and to assemble the necessary data. WSDOT and MDSHA were only able to utilize the website approach through the active participation and enthusiasm of staff at The University of Washington.
- None of the states visited have precise unambiguous location identification and date/time information for their performance, materials and construction data. Performance data are only indicated with the year of the measurements so that it is not known if a major rehab during that year took place before or after the date of the measurements. The performance data in PMS are reported either as an average for an entire mile, or as an average for every tenth of a mile, and for a multi-lane road normally only the right lane is measured. In the construction files records are kept for materials, mix composition, density, etc; these data are linked to (a) certain charge(s) and lot(s) on the road, but it is not recorded where these lots are situated. Another difficulty in relating performance to location is the fact that many QC data are carried out randomly.
- A review of possible data sources and formats shows that the web-based system developed by the University of Washington or a similar system, could be used to store, link and analyze all data needed for a Superpave evaluation.
- All state DOTs have limited personnel in their PMS sections and generally the materials personnel are heavily involved in the design, supervision, QC/QA, and/or laboratory testing of materials. In all cases, these personnel are capable of doing the necessary data collection, but their workload at the present time does not permit it. PMS personnel in general are also fully occupied with other duties.
- All state DOTs visited in this project have an interest and a good potential for assembling the data needed for performance analysis of Superpave projects.
- A well-designed, flexible, operational PMS, and electronic databases for materials and construction, are key elements in setting up operational performance analysis such as examined in this project. In cases of detailed evaluation of aspects of Superpave and/or other complex materials and design concepts, it may be necessary to set up extended PMS subfiles with additional performance measures such as detailed crack type in addition to a crack index typically currently used.
Based on the conclusions and findings of this study and discussions with the sponsors and the five state DOTs visited, the following recommendations are offered:
- It is recommended that FHWA or some other oversight agency explore and undertake a more detailed effort to outline a generic database, which could be applied to develop a nationwide Superpave dataset. This dataset for 5 to 25 states would have broad capabilities for use in analysis of performance for various aspects of Superpave and other pavement concepts and materials.
- The pavement evaluation and measurement protocols previously developed under FHWA auspices (and currently being reviewed for adoption by AASHTO) would provide excellent tools for use in data collection for performance evaluation.
- State DOTs should be encouraged to strengthen their PMS analysis capabilities either with in-house staff or through consultants to use PMS data for engineering analysis. In some cases they should expand their PMS data collection in various subsystems to provide more detailed engineering information.
- It is recommended that collection of data in electronic format be adopted by state DOTs as a means to assemble uniform detailed data collection and electronic storage for Superpave data for exchange and use in evaluation both within the state DOT and nationwide.
- The main requirement for linking databases for performance, materials and construction is to use precise common location identification and date/time information. This identification must be unambiguous over time. Locators can be provided by GPS measurements that are relatively economical and easy to use at the present time. However, they must be tied to traditional location identification information such as project number, mile point, lane, direction, date, etc.
- The integrated website approach developed by The University of Washington has great potential and warrants further consideration.
- FHWA, AASHTO, other national agencies, Universities and the Industry should encourage the preparation of a national plan for use of PMS data for evaluation of Superpave. Each state can use pavement management and related data to evaluate Superpave, but it will be faster and more definitive if several states can work together to set up databases with the required data and combine their efforts to make the necessary performance evaluations. A multi-state effort with a group of 5 to 20 states with coordination among states, can produce a large analysis joint database of lasting value.
- The same approach as indicated in the previous recommendation should be followed for other new materials and design concepts. It would particularly be valuable to outline a plan for future use in evaluating the proposed 2002 AASHTO Pavement Design Guide, which is scheduled to have significant changes from past history but which has not yet been proven by field performance.
- A number of rapid performance testers, such as the Georgia wheel tester, have been introduced for asphalt concrete surfaces. It is recommended that a database element be included in the pavement management database to store this type of data. While the ultimate goal of performance evaluation is long-term performance, it is essential to correlate rapid performance testing devices, heavy load vehicle simulators, and similar test track data into a permanent performance database.
- It is important that data used for performance evaluation be accurately and uniformly collected. Every attempt should be made by state DOTs to continue to codify data collection standards. The data collection protocols developed under FHWA auspices is currently in the process of being finalized by AASHTO and should be adopted for this purpose as soon as practical.
- It is simply good business practice to collect, process, store, retrieve, and analyze data from pavement management and related systems to evaluate new materials concepts, techniques and designs. This is clearly illustrated in this study and by many of the references cited in this report.
Office of Asset Management, Pavement, and Construction