History and need
Managing pavements has become more complex as California must focus on the importance of preserving and maintaining its roadway infrastructure in addition to new construction. To maintain California’s roadways requires having data on the materials used, the surface and subsurface stresses experienced, and conditions and traffic loads.
Many roads within the state system do not have adequate documentation of the pavement structure, especially roads assumed into the state system from counties and cities. In addition, with more roads to maintain, the state must budget resources wisely. To address budgetary allocations and provide the data needed for the pavement design method, Caltrans has begun implementing a new pavement management system called PaveM.
In 1979, California was among the first states to adopt a pavement management system. However, for a long time the system was struggling with a number of problems. The old system was dependent on manual inspection and the data collection was a very slow process for half a million lane Km of paved roads. Besides, the inspection was a biennially process and the data was not as up to date as needed for some analysis. The last drawback of the first established system was that there was no predictive analysis to forecast the future performance level of the system. Finally, in 2000s Caltrans started working with the University of California, Davis, and UC Berkeley to develop the PaveM pavement management system. The goal of the system was to adopt improved ways to collect pavement data to better predict performance, consider what-if scenarios, and invest dollars wisely.
PaveM is Caltrans’ state-of-the-art Pavement Management System. PaveM utilizes pavement performance data to efficiently recommend pavement investments. The system’s approach maximizes the return on investment of every pavement dollar, extends pavement life by treating distresses at the right time, and achieves targeted performance goals. PaveM Tools were designed to aid Caltrans pavement staff in identifying hot spots for needs, focusing on developing right treatments at right times to more effectively plan and deliver pavement projects.
PaveM utilizes pavement history, current pavement condition, current programmed projects, traffic, and climate data to predict future pavement conditions. It recommends pavement repairs needs using optimized strategies within funding constraints. Figure 1 shows different components of the PaveM system.
Figure 2. PaveM DSS components
APCS and GPR
First, the researchers used ground-penetrating radar (GPR) to determine the pavement structure Cost savings were achieved by analyzing only some of the lanes of a roadway, for instance, one side of a two-lane highway or only one or two lanes of a multilane freeway. Caltrans also started an annual process of collecting surface distress data of all pavement in California using an automatic pavement condition survey (APCS) methodology.
The PaveM software was developed using the new distress data. Configuration details include developing the data model on which the system will operate, decision trees that define the appropriate strategies, defining the performance models, and eventually perform optimizations.
The details of the PaveM analytical method has not been publicly shared. However, based the information available on the California office of pavement management website and the PaveM portal and also considering the capabilities of the PaveM the list of analytical method and tools used in the PaveM includes but is not limited to:
• Data mining
• Predicted pavement condition using deterioration models
• A series of decision trees representing different feasible scenarios
• Image processing algorithms to analyze the surface distress
• LCC Analysis
• Unit costs for the treatments
• Optimization model to maximize the network level performance
To develop the PaveM system the research and development team first lunched and immerse database containing data from different data source including the construction and maintenance history of each road segment, pavement type, climate data, traffic data etc. After that the database is fed by the new data generating from the Automated Pavement Condition Survey and Automated Ground Penetrating Radar system. For example the APCS high definition cracking images has been analyzed using image processing algorithms to develop a crack map of the pavement surface.
The PaveM system has many different functionalities and outputs. It helps the Caltrans pavement management office to make better decision regarding their construction and maintenance programs. It also influences most of the design decisions regarding the type of the pavement or the maintenance schedule. However, the main key that makes PaveM a good choice is the scale of the system meaning that PaveM considers the whole network of the pavement over all of the California transportation system and in that it is the first intelligent network level transportation asset management system in Caltrans. Other functionalities of PaveM includes:
• Analyzing the historic and current data to predict the future status of the Pavement performance level
• Life-cycle cost analysis (LCCA) of different pavement preservation strategies.
• Prioritizing pavement treatment plans.
• Supports decision making based on the project optimization tool to propose the right repair treatment at the right time.
• PaveM influences both funding distribution and project selection.
RP List or Recommended Project List, provides users with the list of programmed and PaveM scenario-recommended projects for programming purposes. RP-List Data is data that is exported out of PaveM. This data is based on the fine segmentation of the highway system and contains lane-based data. One can download district-level data, in an Excel (*.xlsx) file, by selecting the district from the PaveM portal.
Pavement Condition Report (PCR)
The PCR tool is one of the main outputs of the PaveM system and is publicly available on PaveM portal. Using the tool one can obtain a spreadsheet containing data from the pavement condition both from the past and future predictions. The tool allows the user to select a specific route in a district and then choose the start and end year of the Pavement Condition Report. The Excel output file will contain the performance level of different part of the route based on the Map21 classification in the three category of good, fair, and poor. The PCR tool is accessible using the link below.