Use of PMS Data for Performance Monitoring with Superpave as an Example

9. Phase 2 - Pathfinder Study in Maryland

9.1.General

The results of the study described in the previous sections were presented and discussed during a project review meeting with several FHWA representatives on February 21, 2001 in Washington DC. Notes of that meeting are given in Appendix E. A preliminary report "Use of PMS Data for Performance Monitoring with Superpave as an Example, Phase 1" was submitted to FHWA on May 30, 2001.

On August 21, 2001, TRDI received a Purchase Order to extend the study with the following statement of work:

"Additional data elements for measuring pavement performance have been identified by TRDI and recommended to be linked with the pavement management database. Based on this recommendation, TRDI will return to Maryland and take information that is currently in flat files or any other inappropriate form and put it in electronic format. TRDI will then electronically link the databases (i.e. materials, PMS, construction, and quality control) in such a way that engineers can readily access the data from all the databases for the purpose of analyzing pavement performance at the network level for long enough periods of time so it can cover a complete life cycle.

Specifically, TRDI will return to Maryland to take data needed to track the performance of Superpave projects and move from flat files to an electronic format and then link the databases (materials, construction, QC and pavement management) as an example to show the states how this should be done. Only one state, Maryland, will be revisited and a report written to document the process".

On August 27 and 28, 2001, the Project Team met with representatives of Maryland SHA to discuss the following:

1. Database to be used for linking performance/materials properties/construction details (mix design and testing, construction, performance, maintenance)
2. Amount of Data Transfer for Path Finder Study (period, types & number of projects)
3. Format of electronic database (AASHTO Superpave, Excel, Web-site)
4. Process of Data Linking (including analysis and reports)
5. Efficiency and Reliability of Data Collection, Transfer and Linking process (how much time and how much detail are needed)
6. Example(s) of actual analysis of Superpave performance (in final report)
7. Actual staff arrangements

A report about this meeting can be found in Appendix G-1.

On October 29, 2001, members of the Project Team had a follow-up meeting with representatives of the Office of Materials and Technology of the State Highway Administration of Maryland (MDSHA) at their office in Maryland. The following points were discussed:

1. The progress made so far in identifying data needed for analysis and evaluation purposes of materials and construction techniques, and the selection of data that are considered important for the project.
2. The progress made by Maryland SHA in using the web site approach for SMA mixes in cooperation with the University of Washington.
3. Agreement on action points.

MDSHA prepared a list with proposed data fields, data availability and comments for QC/QA data, Mix design information, Pavement design information, and Pavement Management data, together with a suggested list of reports that should cover the effects of various parameters on rutting, reflective cracking, ride quality and cracking. The Project Team proposed to add a few data fields in the areas of existing pavement structure, mix temperatures during spreading and compaction, performance data of laboratory- and actual mixes, and date when pavement section is opened to traffic. More detailed Notes about this meeting are given in Appendix G-2.

The PMS does not yet contain data on maintenance, but it was agreed that the linking system should provide for the inclusion of maintenance activities in the future.

The recommended data fields are given in Table 2, and Table 3 lists the types of analysis and evaluation that would be desirable.

Table 2 Recommended Data Fields to Include in Pilot Study
Key to Data Type Column: E=Electronic, PF=Project File (paper), N/A=Not available

Data Field	Data Type	QC/QA	Mix Design	Pavem. Design	PMS
AADT	E
Aggregate Consensus Properties	PF
Aggregate Source	PF
Aggregate Type	PF
Ambient Temperature	PF
Asphalt Content	E
Asphalt Producer	E
Binder Test Results	E
Binder Type	E
Compaction Level (N Design)	E
Cracking Condition	E
Cumulative ESALs	PF
Daily Paving Location (location of lots)	PF
Date when pavement section was opened to traffic	N/A
Density (field compaction) from cores & nuc gauge	E
Density at N Max	PF
Design Constraints	PF
Design Life	PF
Existing pavement layer thicknesses	PF
Friction Condition	E
Gradation	E
HMA Unit Price	PF
Layer Type	E				4
Load Spectra	N/A
Location of Project	E
Longitudinal Joint Condition	N/A
Max Gravity	E
Milling/Grinding Recommended	PF
Mix Design Number	E
Mix Pay Factor	E
Mix Temperature prior to compaction	PF
Mix Temperature after compaction	PF
Mix Type	E
No and Type of Layers Recommended	PF
Patching Recommended	PF
Pavement Type	E
Paver Type	PF
Paving Contractor Name	PF
Paving Time of Day (Day or Night)	PF
Performance Test Results	N/A
Pictures	N/A
Pre-overlay Condition	PF
QA Sample Location	PF
RAP Quality	PF
RAP Quantity	E
Rating of Joint Construction	N/A
Rating of Segregation	N/A
Raveling Condition	N/A
Reflective Cracking Condition	N/A
Ride (profilograph results, profiler results in 2002)	PF
Ride Condition	E
Ride Pay Factor	PF
Rolling Pattern	N/A
Rutting Condition	E
Section (open or closed, manholes, entrances)	PF
Soil Strength	PF
Soil Type	E
Special Features (joint tape, saw seal, etc.)	PF
Structural Capacity	PF
T-283	PF
Thickness of new layer(s)	N/A
Tonnage	PF
Transfer Device (Y/N)	PF
Type of Improvement Designed	PF
Type of Rollers	PF
Volumetrics	E
Weather Conditions	PF
Wedge and Level Recommended	PF
Why Binder was Bumped	PF

9.2.Collecting Data and Conversion to Electronic Format

In Maryland only the Superpave projects designed after June 1999 have sufficient data to make the performance analysis meaningful. MDSHA therefore selected 7 Superpave projects carried out since 1999, and made an effort to collect all available data listed in Table 2 from their QA/QC, Mix design, Pavement design and Pavement management records, and to import these data into an electronic database. Apart from the data that were not available (N/A in Table 2) they were unable to retrieve data on Aggregate Properties and on Binder Test Results. Other data, mostly those in paper files, were hard to retrieve. An overview of the sources of data and the difficulty to retrieve them, is given in Table 4. More details are given in Appendix G-4.6. For the seven Superpave projects it took 2 staff members about 4 weeks working full time to retrieve the data and to enter them into an electronic database. They used an Access database about 45 MB in size. Electronic data storage will in the future improve this process and speed it up considerably.

9.3.Setting up Database in modified UW Web-site System

As mentioned in Section 8.6. the University of Washington (UW), in cooperation with Washington State DOT and NCAT, recently developed a fully integrated website that contains relevant data of Superpave contracts, including performance data [White 02]. This new development was possible because nearly all essential data on materials and construction are available electronically. In addition a major effort has been made to link these data to performance measurements from WSDOT's Pavement Management System (PMS). The data is available on a website (see Appendix F) from which they can be easily organized, downloaded and analyzed.

In the initial phase of the study the Project Team considered the use of a modified version of the QC/QA module of the Superpave/AASHTOware data set [AASHTO 00] for collecting and storing the required Superpave data. After further review and knowledge of the University of Washington system, two reasons emerged to support why it proposed to use the web based system developed by UW for the Maryland Database with Superpave projects:

1. AASHTO recently decided to stop promoting and servicing their Superpave/AASHTOware data set package.
2. Several of the data fields listed in Table 2 are not included in this AASHTOware package.

Dr. Joe Mahoney and Mr. George White of the Civil & Environmental Engineering Department of UW offered their help in making their web-site system available for the seven Superpave projects from MDSHA. In order to do this George White undertook the following:

• Modify and extend the system to:
  • Accommodate all data fields given in Table 2
  • Extend the functionality of the system to include the analyses and plots listed in Table 3.
• Checking the MDSHA database and entering the data in the web-site system.

The total effort of George White for the above mentioned activities was about one man-month. Most of this effort went into the last point, clarifying the meaning and inter-relationship of the data, getting additional data, adjusting units, converting systems, etc.

UW received most of the data from MDSHA during the first half of December, 2001. On December 18, 2001, two members of the Project Team visited UW to discuss the possibilities for a life demo of the capabilities of the UW web-site system using the Maryland data during the TRB 2002 Conference. (A visit report is given in Appendix G-3). It appeared that MDSHA had forwarded a substantial database with complete data for most of the fields. Still, for 10 fields no data were available at all, and for several other fields there were insufficient data. Consequently no analysis could be done for cracking or other distresses, for the effects of mix temperature, the influence of the use of a MTV, the influence of day or night paving, the effect of different roller patterns or roller types, or the effects of actual versus designed layer thickness. It appeared that data with exact locations of lots and sub-lots were also missing, but it was agreed for one section only to divide the lots evenly over that section so that, for demonstration purposes only, the correlation between a lot and its performance could be shown. Another limitation of the data for the seven Superpave projects is that there are only one or two years of performance data, so it is difficult to create meaningful plots of performance over time. Some of the older Superpave projects do have several years of performance (see Appendix D), but these projects have insufficient data on materials and construction.

9.4.Demonstration of Capabilities of Web-site System of UW

The final presentation of the current project took place on January 15, 2002 at the TRB Conference in Washington DC. The purpose of the meeting was to give the FHWA and representatives of industry, universities, committees and state DOTs an overview of the progress made and a demonstration of the web-based evaluation system developed by the University of Washington. Representatives of the FHWA, TRB, AASHTO, Maryland SHA, Washington State DOT, NAPA, Asphalt Institute, NCAT, Battelle and the Universities of Maryland and Washington attended the meeting.

After a welcome and introduction by Carl Monismith, Pim Visser gave an overview of the project and of the progress made so far. He described the concept of linking materials and construction data to performance and distress measurements; the possibilities in case of an ideal situation of having all required data available in electronic format, and the limitations of the current situation where many data are either not available or difficult to access. He reviewed phase 2 of the project: the Pathfinder Study carried out with data from Maryland SHA, with as two main elements the actual collection of data, and their incorporation into the web based system developed by the University of Washington (UW). See also Appendix G-4.

Pete Stephanos described the efforts of MDSHA to collect the agreed set of data. For each data category he described the source of the data, and the level of difficulty to retrieve them. See also Appendix G-4.

Joe Mahoney introduced the development and capabilities of the HMA electronic web-based database and evaluation system, used by the State Pavement Technology Consortium of Texas, Minnesota, California and Washington on an experimental basis.

George White presented a "live" demonstration of the web-based evaluation system developed by him in cooperation with WSDOT and NCAT at the University of Washington. He showed two websites: http://hotmix.ce.washington.edu/hma/ with the recent data from MD and http://hotmix.ce.washington.edu with data from WA, TX and MI. Since many data were still missing it was difficult to show well-defined plots, however most of the desired relationships were shown and the audience got a good idea of the capabilities of the system.

It was stressed that the demo should be judged on capabilities, rather than these interim results. Additional information about the websites, together with examples of web pages, is given in Appendix F.

The animated discussion after the meeting can be summarized as follows:

• It was clear that the system is very flexible, and that it is able to accommodate the different requirements of various agencies, but that there will be a need for some degree of consistency for data and the use of these data to make broader comparisons possible. One additional advantage of the system is that the information can be refreshed and updated at any moment, so it is possible to work on the basis of the latest available information. The use of "real time" QC/QA data during construction could be very beneficial.
• The FHWA is very interested to pursue the presented concepts, for instance by assisting with a Pooled Fund study in which various states can participate, but in the short term there is a lack of funds,
• Several representatives agreed that the concept would be a very useful tool to secure required pavement performance data for use in the AASHTO 2002 Pavement Design Guide and to evaluate the performance of pavements designed with this Guide,
• Industry representatives indicated their willingness to cooperate by promoting the use of data in electronic format. They considered it advantageous to have one complete set of data that could be used by all parties,
• MDSHA, WSDOT and UW want to continue with this system, independent of the possible start of a multi-state pooled fund study, but they are hoping for some contributions in the short term from some DOTs and/or the industry,
• All participants welcomed and approved of the concept, and it was realized that it is important now to capitalize on the considerable momentum that has been built up by the current promoters of the concept. Some mechanism should be set up and funded quickly to continue the good work, and to widen the promotion to other states.

More details about this Special Meeting can be found in Appendix G-4.