1.0 Project Background
Based on the results of Phase 1, this section presents a detailed work plan for Phase 2. This work plan clarifies and expands upon the Phase 2 work plan presented in the Task Order Proposal Request and the project team's original proposal to reflect the decisions made during Phase 1.
Phase 2 consists of the following tasks (Phase 1 consisted of Tasks 1-4):
- Task 5 - Data Collection
- Task 6 - Identification of Critical Data and Tools
- Task 7 - Pilot Study Reports
- Task 8 - TWG Meeting
The following sections detail the specific work steps and deliverables to be provided for each of the tasks included in Phase 2.
3.1 TASK5 - DATA COLLECTION
Objective: Collect and gather pavement and obtain bridge condition data on a three-state pilot study corridor, compare it to and analyze it along with State provided data, HPMS, and NBI data.
The selected pilot study corridor is I-90 in South Dakota, Minnesota, and Wisconsin as shown previously. This is approximately an 864-mile corridor.
A data gathering, collection and analysis plan will be developed based upon the metrics recommended during Phase 1 and will be submitted to the FHWA for review and approval prior to undertaking any further activity under the task. Upon approval, we will begin the data collection and gathering process.
In order to undertake this process, the team will obtain data from three sources, including:
- National Data;
- State Data; and
- Field Data.
The study team will access data from the National Bridge Inventory (NBI) for the selected corridor. These data will be examined to identify bridges on the corridor and identify the type and condition of each bridge.
The study team also will access data from the Highway Performance Monitoring System (HPMS) to obtain inventory and pavement information from HPMS sample sections that reside on the study corridor. The data will be formatted and input into the project database for further evaluation in later stages of the project.
The State data gathering phase will involve gathering of data that resides in the States' management systems.
The team will establish contact points with each of the State DOTs (South Dakota, Minnesota, and Wisconsin). We intend to utilize these contact points as the conduit for the team to perform the State data gathering effort for pavement and bridge information.
One of the key success factors for this task will be to manage our interaction with the States during the data gathering process and make it efficient and painless for the participating States. This management strategy will include assigning one person to act as the project team's point of contact with each State (our State POC). This will negate the possibility of several team members contacting the State asking for the same information multiple times. In our experience, States are receptive to providing data. However, it is a burden in time and resources to perform this activity and not productive to have more than one point of contact. Our approach must be efficient by providing very clear instructions as to the data and associated documentation needed for the study and clear expectations for delivery must be established early in the process. When follow-up is needed, our State POC will perform the follow-up to ensure consistency in message and request.
The team will develop a data request that includes the type of data requested, the required data elements, desired format, supporting information (standards for data collection, frequency, sampling interval, etc.) and deadline for receipt. The data request will be submitted to the State liaison and a follow-up call will be made within two days to determine if the State has any questions or concerns with the data request. In order to reduce the effort by the State, it is likely that we will request the data in its native format (with appropriate metadata) so that the State does not need to reformat their data. The data elements to be requested from each State include:
- Inventory and linear referencing data including right-of-way images (if feasible/available);
- Financial information (if available);
- Traffic data (AADT, and percent trucks);
- National Bridge Inventory (NBI) data;
- Pavement type (flexible, composite, PCC);
- HPMS 2010+ pavement data (from the State's HPMS submission); and
- Pavement condition information from the State's Pavement Management System (both directions) including:
- Pavement structure (best available);
- PCI data (if available);
- Pavement maintenance/rehabilitation history including costs (best available);
- Falling Weight Deflectometer data (if available);
- Cracking/Distress (in raw and processed format);
- IRI (raw and processed);
- Rutting (raw and processed); and
- Faulting (raw and processed).
Upon receipt of the data, it will be checked for completeness and outlier data and anomalies will be discussed with the State POC. The data will be entered into the project database in a consistent format across the three States so that it can be compared during the analysis phase of the project.
A critical task for this project will be to collect data in the field in order to compare apples-to-apples on a multi-state network. As decided with FHWA, bridge information will not be collected in the field as the project team will rely on NBI data provided by the DOTs and obtained from FHWA.
Previous projects have shown that while each State collects data, even "standard" data such as IRI can be collected using different methods, protocols, and equipment resulting in incomparable outputs. The results of this task will allow a comparison between network performance and health measured using consistent procedures and equipment, health measured using State-reported HPMS and NBI data, and additional condition data contained within agency pavement and bridge management systems. Field data collection will focus on pavement condition information.
To perform this task, the team will use the following multi-step process:
Pavement condition information will be collected using a Rolling Wheel Deflectometer and an automated multi-sensor pavement data collection vehicle moving at highway speeds. For pavement data collection, the project team will focus on field data collection of the following information:
- Rolling Wheel Deflectometer (RWD) data4;
- Right-of-Way (ROW) digital images (used for QA purposes); and
- Pavement condition data, including:
- Cracking/Distress (in raw, HPMS 2010+, and ASTM Pavement Condition Index - PCI - format5);
- IRI (raw and processed using ProVal);
- Rutting (raw and processed using AASHTO Provisional Standards); and
- Faulting (raw and processed using ProVal).
The specifics of data collection will be outlined in the detailed data collection plan described earlier. The data collection plan will include specifications for equipment (including calibration requirements), data types, data collection intervals, and interpretation and analysis parameters. In addition, data quality parameters will be established to be used to check the completeness and reasonableness of the data collected. Data formatting and storage requirements will also be addressed. The plan will contain the segments to be examined, State coordination requirements, and safety considerations. This data collection plan will be shared with FHWA and the TWG prior to field data collection.
Once the data collection parameters have been established, the team will secure data collection vendors. The same data collection vendor and equipment will be used in all three pilot States. The FHWA has stated that they expect RWD testing to be performed on a sample of the network. Applied Research Associates, Inc. (ARA) is the only vendor who can perform this type of data collection; thus the team will subcontract with ARA to perform the RWD data collection. Based on preliminary cost estimates provided by ARA (~$60/mile = ~$52,000) it should be possible to conduct RWD testing of the entire corridor (in the east direction so as to be consistent with HPMS 2010+) using project funds. We will coordinate the sample size (whether full or partial coverage) with FHWA prior to execution of the RWD testing.
The team will also contract with a data collection vendor to perform the other pavement data collection activities. We expect that the vendor will perform a 100 percent survey in the right-most travel lane of the corridor in one direction (in the east direction so as to be consistent with HPMS 2010+) and that we will establish a sampling procedure to develop cracking indices from this data so as to minimize data collection/processing costs (it is estimated - rough estimate - that data collection and data processing will cost approximately $160 per mile = ~$163,000). The sampling interval will be developed as part of the data collection plan. It is expected that ROW imagery, IRI, faulting, and rutting will be collected for the entire corridor.
4According to the vendor, texturing and tining in PCC pavements renders RWD data unreliable, thus only results for asphalt concrete data will be analyzed.
5It should be noted that automated data collection procedures may not be able to recognize some distresses (e.g. raveling and others) used in the PCI calculation without extensive manual intervention.
Once the data collection types, procedures and vendors have been identified, data collection can commence. One of the key parameters to be addressed during this step is to engage, inform and coordinate with the State DOTs involved with the corridor study. Our State POC will inform the contact at each DOT of the timing, location, and type of data collection being performed. Since all of the proposed data collection is performed at highway speeds, traffic control is not required.
Data collection will commence according to the plan developed under step 1. It should be noted that several weeks will be required after data collection for the vendor to process the RWD and pavement condition data. It is expected that 200 miles of either RWD or pavement condition data can be collected in a day so the entire corridor can be collected in a one-week period assuming weather is not a factor.
For pavement condition data, the selected data collection vendor will provide the project team with raw and processed data in accordance with the data collection plan. The project team will perform quality assurance review of the collected data to determine completeness, reasonableness, and calculation accuracy.
Progress in field data collection will be reported to FHWA on a weekly basis by email and a bi-weekly teleconference will be held to discuss progress.
The team will review and identify issues related to data collection, uniformity, and availability. In particular we will investigate to what extent independent (but uniform) data gathering and collection identifies differences between national data, State data and the collection sample. We will also investigate data consistency between national level data and State supplied data for the same information (for example NBI data from the FHWA versus State supplied NBI data, and national HPMS data versus State supplied HPMS data).
This comparison will be accomplished by calculating and comparing key statistics (e.g., minimum, maximum, average and standard deviation) for each data set. Identifying the existing limitations of each method of data collection and data sources will be a key finding from this study.
Another key analysis parameter will be the comparison of various types of pavement distress indicators. For example the HPMS pavement condition information yields a list of values for all of the data items. However, there is no method in existence to convert these values into a condition index (such as the ASTM PCI). By necessity the team will need to create such a transfer function and ground-truth the results using information collected as part of the study. One method we will explore is to emulate the PCI calculation procedure in structure. Simply put, the PCI procedure uses a series of deducts based on the extent and severity of each distress. These deduct curves are used to assign a deduct which is combined for all distresses and subtracted from 100. We intend to explore this method to convert raw data to a condition score. The PCI data collected during the pilot can be used to validate or correlate the HPMS "PCI" and the ASTM based PCI. We will also consider the use of the RWD structural data6in the HPMS PCI calculation or as an adjunct to the distress/IRI based HPMS data. Pavement performance will be assessed for both directions using PMS data and will be compared to pavement performance based on HPMS data and data collected by the study team (which will be collected in one direction). The purpose is to understand the degree to which performance based on data from one direction represents the performance of the entire system.
In addition to the above, an algorithm for a new NBI-based measure of structural adequacy may also be developed.
The results of this review will be reported under Task 7.
- Data Gathering, Collection and Analysis Plan
6Usually this is done using a simple "strong versus weak" classification.
Objective: Identify data and analysis tools needed to develop the infrastructure performance and health reports in Task 7. The project team will review the outcomes from the TWG meeting held in Task 3 as well as the results of the data gathering effort in Task 5 to identify missing data and analysis tools required to produce the pilot study reports in Task 7. Depending on the specific metrics selected for the pilot study and the data provided by the pilot States, the team may uncover issues with the availability and/or quality of the data and the availability of analytical models (or tools that encapsulate these models). For example, the team may determine that additional data will be required from all States in order to fully implement a national standard for pavement or bridge health. However, States may lack equipment, expertise and/or funding to collect and report these data. Also, given variations in materials and construction techniques for roads and bridges, there may be issues with capturing consistent data in all cases. It will be especially important to review data quality and data consistency among the pilot States.
For each deficiency, the team will propose one or more resolutions along with estimated costs. These costs will be developed in consultation with the pilot States regarding the level of effort to obtain these data.
In situations where analytical capabilities are not available, the team will draw on its own experience in developing pavement and bridge management tools as well as condition and performance models to estimate the cost of designing and implementing tools at different levels of sophistication. The team also may consult with FHWA personnel to gain a better understanding of the environment in which these tools might be deployed.
At the conclusion of this task, the team project manager will submit a memorandum to the task monitor. This memorandum will document any gaps in data or analytical capabilities as well as potential resolutions and estimated costs for each.
- Memorandum detailing gaps in available data and tools and recommended strategies, and estimated costs, to address these gaps.
3.3 TASK 7 - PILOT STUDY REPORTS
Objective: Develop sample Infrastructure Performance and Health reports7. Compare the output of these reports with other pavement and bridge reporting mechanisms. Document results to date, including recommendations on how to address deficiencies in the data gathering process.
Following analysis of pavement and bridge data from the pilot States, the project team will prepare sample reports using the data obtained in Task 5. The specific metrics, visualization techniques and presentation medium used to show both performance and health will be based on feedback received from FHWA and the TWG supplemented by the team's experience in preparing and presenting this type of information.
At the present time, we envision that these reports will share a common structure. Users will be able to specify a geographic area (e.g., State, corridor, region, national). This will constrain the assets included in the report and determine the level at which information will be summarized. Each report will be divided into separate sections. Potential sections include:
- An overview of condition and health metrics in tabular, graphical, and static map formats, summarized by one or more categories (e.g., good/fair/poor, asset size, functional class); and
- A critical asset list that identifies infrastructure (either individual items or specific regions) in need of immediate attention.
Each report will be designed to operate off different data sets (e.g., State data, national data, pilot study data) to the greatest extent possible. It is possible that the appropriate data may not be available in each data set. Task 6 will address these specific deficiencies.
The team also will consider how to display condition and health information in a dynamic mapped-based format using a Geographic Information System (GIS)8. Our team has extensive practical experience using GIS to show asset data, including development of a system for FHWA to display pavement and bridge metrics for the Mid-Atlantic States using the Interstate 95 Corridor Coalition's Integrated Corridor Analysis Tool (ICAT). ICAT provides a consistent road network and an Internet-based mapping system for the I-95 corridor. We anticipate development of a prototype system equivalent to ICAT using road networks provided by the pilot States. We will also review dashboards from the three pilot States to learn how they report their performance measures.
Throughout the report design and creation process, the team will schedule two or three web conferences at which results to date will be presented to FHWA and TWG personnel. Comments received on these interim results will be used by the team to adjust the content and layout of the reports to ensure that the final product accurately reflects the vision of all stakeholders. These web conferences will include a discussion of data requirements, techniques used to create the reports, and any other issues that may affect the adoption of these reports at a national level.
Once the sample reports are complete, the team will hold a final web conference to allow FHWA and TWG members to provide any additional comments. At this time, the team will discuss not only the process for creating the reports but what the reports indicate with regard to infrastructure condition and health in the pilot States. Following this conference, the team will prepare a draft report that documents:
- Information on data obtained, including issues of data quality and consistency;
- Details on the sample Infrastructure Performance and Health reports, including what these reports indicate for the pilot States and any caveats regarding these results;
- Recommendations on how to improve the data gathering and analysis processes to the point where they can support the infrastructure reports;
- Recommendations for State DOTs for explaining any differences between FHWA's infrastructure health index and their own approach;
- Manual versus automated crack detection and reporting;
- Recommendations for future HPMS development including a "wish list" of improvements to be considered during the next update; and
- Estimated level of effort to roll-out these reports on a national level.
The team project manager will submit the draft report in electronic format to the task monitor, who will distribute it to FHWA, TWG, and pilot State personnel for comment. The task monitor will gather any comments and submit them, in writing, to the team project manager. Once all comments have been received, the team will prepare and submit a final report, in electronic and hardcopy formats, that incorporates all the comments received.
- Sample Infrastructure Performance Report;
- Sample Infrastructure Health Report;
- Draft report (electronic); and
- Final report (electronic and hardcopy) that incorporates comments from FHWA, the TWG and the pilot States.
7The project team will coordinate with FHWA so as not to duplicate other ongoing FHWA efforts (e.g. HIPAT, QlikView).
8 The GIS work and any tool developed as part of this effort will be considered a prototype that illustrates a "proof of concept" of the approach. Additional work will be necessary to determine how best to deploy a production version of this system that, for example, enables FHWA to distribute the information widely over the internet without licensing issues.
3.4 TASK 8 - TWG MEETING
Objective: Present infrastructure reports to the TWG and discuss the viability of using these reports on a national scale.
Following review of the Task 7 deliverables by FHWA and TWG personnel, the project team will organize a one-day meeting to discuss the Infrastructure Performance and Health Reports and any issues related to using these reports at a State, corridor, regional or national level. The team will lead meeting participants in a discussion of issues such as:
- What the reports are intended to show;
- What the reports do not show;
- Impediments, if any, to implementing the reports using current data and analytical methods; and
- How the reports can be used to understand infrastructure performance and health.
The TWG meeting will be held at a location to be determined in consultation with FHWA. The team will assist TWG and pilot State participants with travel arrangements, if necessary, and reimburse all travel costs. Discussion materials will be distributed in electronic form at least one week prior to the meeting date.
- One-day TWG meeting (including meeting materials and travel support for up to six TWG members and, if appropriate, up to six participants from the pilot States).
Summary of Deliverables
As a result of the activities of Phase 2, the following deliverables will be developed by the project team:
Table 3-1: Summary of Deliverables
|Task 5 - Data Collection||Data Gathering, Collection and Analysis Plan|
|Task 6 - Identification of Critical Data and Tools||Memorandum Summarizing Gaps in Available Data and Tools|
Task 7 - Pilot Study Reports
Sample Infrastructure Performance Report
Sample Infrastructure Health Report
Draft Project Report
Final Project Report
|Task 8 - TWG Meeting||
One-day TWG Meeting
Figure 3.1 contains the proposed schedule for execution of Phase 2, shown in context with the entire project schedule. We propose to follow the schedule developed during the original proposal process.
Figure 3.1 Phase 2 Schedule