FHWA kicks off an inventory program for roadway support structures to assist management efforts at the National Park Service.

The U.S. Department of the Interior's National Park Service (NPS) is responsible for managing and maintaining nearly 5,000 miles (8,047 kilometers) of paved roads and parkways across more than 250 properties nationwide. Many of the scenic roads traversing the country's national parks were constructed by the Civilian Conservation Corps during the 1930s and are beginning to show their age. So too are the retaining walls and other support structures that secure, protect, uphold, carry, or otherwise convey these roads across challenging terrains. Unsung heroes among roadway infrastructure, these walls and other structural elements hold historic significance for NPS, and many have been preserved and rehabilitated over the years to meet modern traffic and safety requirements.

Like the roads themselves, numerous subsidiary roadway features, such as retaining walls, bridges, culverts, and traffic barriers, are critical assets that require careful maintenance and rehabilitation. Referred to as "equipment" in asset management parlance, these features are major contributors to the safety and accessibility of the NPS road system and represent substantial investments in roadway infrastructure. In addition to the pavement assets, NPS is responsible for appraising and managing the maintenance needs of all assets associated with the roads, including retaining walls and other equipment. Given the wide range of geographic settings and extensive public use of NPS roads, defining the backlog of roadway equipment needs in terms of location, quantity, condition, and failure consequences represents a major challenge for NPS.

Currently, NPS and the Federal Highway Administration's (FHWA) Office of Federal Lands Highway (FLH) collaborate to assess park roadways and bridges using two inventory programs: the Road Inventory Program (RIP) and Bridge Inspection Program (BIP). In 2005, the Park Facility Management Division of the NPS Washington Support Office commissioned FLH to develop an inventory program for retaining walls, similar in scope to the ongoing RIP and BIP inventories. The mission of the wall inventory program (WIP) is to define and quantify wall assets associated with park roadways in terms of their location, geometry, construction attributes, condition, failure consequence, cultural value, apparent design criteria, and cost of structure maintenance, repair, or replacement. The wall inventory will provide data to the RIP to update equipment assets associated with the parent roadway asset. NPS asset managers will periodically reassess retaining walls in the parks participating in the WIP to ensure that timely, accurate information is available to support NPS asset management initiatives.

The WIP data will feed into the Facility Management Software System, the existing data hub that NPS uses to document, manage, and plan efforts related to park assets. In this data management system, condition assessments for roadway and bridge structures are expressed as deferred maintenance costs, which are then divided by current year replacement costs to arrive at a facility condition index (FCI). According to the NPS report Budget Justifications and Performance Information for fiscal year (FY) 2006, "to ensure that its capital asset investments are made as efficiently as possible, the NPS is incorporating FCI analysis into the prioritization process by comparing the existing FCI of a facility against the proposed FCI after the construction investment." Coupling this condition prioritization with an asset priority index, which measures the facility's importance to a park's mission, NPS managers can make more informed decisions on capital investments, focusing maintenance and construction priorities on value rather than cost alone.

Developing the WIP has been a multiphase effort, with FLH division office and NPS personnel sharing responsibility for each task. The phases included researching existing State inventory programs, developing a customized wall inventory and assessment system suited to NPS asset management requirements, and piloting the methodology at several national parks. FLH took the lead on performing field inventories, while NPS handled integrating the WIP data into its data management system. Since the program launched in 2005, the agencies have completed wall inventories capturing condition assessments and deferred maintenance costs for nearly 3,500 retaining walls within the WIP database.

Over time, "the WIP will benefit NPS in several ways," says WIP Project Manager Dave Keough, with the Park Facility Management Division, NPS Washington Office. Having developed the WIP, "we have credible documentation of the total value of roadway retaining walls, as well as the cost needed to repair and maintain them. This goes a long way when trying to justify budget requests. It also provides solid information for various park units to prepare sound project proposals. When future projects are in development, this information can save money and improve designs."

Program Conception

Based on their experience with the BIP, the WIP development team decided that the WIP should be more than just an inventory program, says John R. Thiel, P.E., bridge management team leader with FHWA's Eastern Federal Lands Highway Division. "The BIP is essentially a safety program, following the National Bridge Inspection Standards that were written into law after bridge collapses had caused loss of life. BIP assesses the condition of each structure and develops recommendations and cost estimates for corrective actions. A bridge management system then utilizes these data and other tools, such as priority indices, to systematically develop and prioritize bridge improvement projects. Based in part on the BIP and bridge management system models, the WIP utilizes a database that stores information provided by inspectors regarding wall conditions, needs, and indices that establish risk and priority for repair of each wall."

In 2005-2006, FLH investigated the feasibility of developing and conducting retaining wall inventories for NPS. Preliminary efforts provided early recommendations for inventory methods and practices supporting the needs of the NPS data management system. More specifically, initial efforts concentrated on the following six tasks: (1) a state-of-the-practice review; (2) identification of the range of wall types and components, and estimation of costs for wall rehabilitation, repair, and replacement; (3) determination of inventory size and breadth; (4) development of a plan to collect wall data that is consistent with existing FLH bridge and road inventory programs; (5) development of a methodology for assessing the condition of wall components; and (6) development of procedures to document cultural resource issues.

Although each of these tasks helped define the current inventory program, FLH officials cite their review of practices underway within Federal, State, and local agencies as the most influential factors in developing the WIP for NPS. FLH canvassed 23 State departments of transportation (DOTs), partner land management agencies, and several major municipalities for experiences with retaining wall inventories. Most of the surveyed agencies reported having bridge and/or roadway inventories tied to an asset management system, but few had moved beyond acknowledging the need to incorporate retaining walls into their asset programs. In fact, only seven DOTs and one municipality had any substantive experience with identifying and inventorying wall assets. In most of these cases, the inventories are limited to simple cataloging systems tied to existing bridge or roadway infrastructure surveys, include only new walls—that is ignore the backlog of existing structures—or focus only on one particular wall type, such as mechanically stabilized earth (MSE) structures.

Prior to WIP development, the Washington State Department of Transportation and Colorado Department of Transportation both had developed extensive wall inventory and assessment methodologies but had yet to implement their programs fully, primarily due to the associated high costs of performing field assessments on thousands of walls statewide. Of the transportation authorities canvassed, as of 2006, the Oregon Department of Transportation (ODOT) was the only one to actually implement a statewide asset management system. Under the planned auspices of an Office of Asset Management, ODOT's program integrates some 62 inventory components spanning structure assessments, financial elements, and asset planning.

In addition to State DOT efforts, the city of Cincinnati has a formal wall inventory system in place, largely modeled after its bridge inspection and inventory program. As of 2006, the city had inventoried approximately 1,800 retaining walls, representing nearly 60 miles (97 kilometers) of wall length and more than $170 million in current year replacement value.

The preliminary review of existing wall inventory efforts underway within the transportation community resulted in several key findings that guided development of the WIP for NPS:

Program cost. Streamlining data collection processes and procedures can minimize costs. The tendency to develop broad, all-encompassing engineering assessments often results in impressive data collection schemes on paper that only later are determined to be prohibitively expensive to implement on the ground.

System integration. Stand-alone inventory systems can greatly simplify processes during system development, but a common database architecture and software platform is critical to future integration with other asset management efforts.

Qualified inspection. Accurate asset data underpins an effective asset management program. The collection of high-quality data on wall conditions is directly related to the careful definition of wall attributes and elements, plus training qualified field inspectors.

WIP Development

Following conceptual development of the WIP in 2006, FLH and NPS team members defined approximately 65 wall attributes that should be logged, measured, calculated, or assessed during field inventories. They also developed procedures, forms, and associated guides and cost information for collecting data in the field, and then created a Microsoft Access®-based, searchable WIP database. The team also conducted several pilot studies in the summer and fall of 2006, for example, at Sequoia National Park and Crater Lake National Park.

"The biggest surprise for me was just how difficult it is to consistently define, measure, and categorize all the different kinds of walls we encountered," Keough says. "The condition assessment, which would seem to be the more difficult task, was often easier than just figuring out what kind of wall the inspector was looking at."

In some circumstances, for example, inventory teams can encounter difficulties in classifying a particular wall's function or in determining whether a structure qualifies for inclusion in the inventory at all. Is the wall on the inside of a switchback, a fill wall, or a cut wall? Should it be considered a wall with a culvert, or a culvert headwall? Is it an integral part of the bridge wingwall and, therefore, covered under the BIP, or does it primarily support the bridge approach? Is it a parapet extending above a wall, or is once-retained-earth missing from the top of the wall? During development of the WIP, the inventory team often faced the challenge of describing these types of walls. Further still, the wall inventories represented only an initial screening of wall asset needs for a given park. More detailed assessments will be necessary before the parks can begin programming wall repairs or complete structure replacements.

The inventory team used five general data categories to describe measure, evaluate, and rate wall attributes and to define and quantify WIP assets.

Wall location. The team identified the location of walls by park name, route number and name, side of roadway, start and end mile points for the wall, and latitude and longitude of start of the wall.

Wall description. Walls are described by function, type, year built, architectural facings, and surface treatments. The team recorded measurements for wall length, maximum height, face area, face angle, and vertical and horizontal offsets from the roadway. Inspectors photographed each wall, noting location relative to the roadway, major wall features, and overall element conditions.

Wall condition assessment. Inspectors described conditions of the primary and secondary wall elements relative to the extent, severity, and urgency of observable distresses, and then rated them numerically, giving due consideration to data reliability. They also evaluated and rated overall performance of the wall system (global performance of the entire wall system), with all ratings weighted and combined to arrive at a final, overall wall condition rating.

Wall action assessment. To determine a recommended action, the team considered the following: (1) the final numerical condition rating of the wall element, (2) any identified requirements for further site investigations (measure of data reliability), (3) the apparent design criteria employed (such as American Association of State Highway and Transportation Officials criteria), (4) any cultural concerns (such as a historic or a context sensitive wall), and (5) the consequences of wall failure. Potential actions included no action/monitor the wall, conduct maintenance work, repair wall elements, replace wall elements, and replace the entire wall.

Work order development. Where work orders are needed, the team provided brief, yet descriptive work orders to outline maintenance, repair, or replacement actions. They generated unit costs for major work items from the WIP Cost Guide (a compilation of cost data for repair and replacement) and/or available cost data from the park to arrive at preliminary estimates of cumulative deferred maintenance.

Of the 23 primary and secondary wall elements defined in the WIP, inspectors in the field needed to describe only those that were applicable to the particular site (generally 5-15). The resulting concise, written narratives characterized the severity, extent, and urgency of element distresses. Inspectors described wall conditions within four general distress categories: corrosion/weathering, cracking/breaking, distortion/deflection, and lost bearing/missing elements. Then they determined the condition ratings by applying a rating scale that ranges from 1 to 10, with 10 being best condition and
1 being worst.

The requirement for sound engineering judgment in the WIP is most apparent in the manner in which inspectors recommended wall actions. Whereas similar condition-based inventory systems might directly correlate a numerical rating to a specific action, the WIP assessment methodology provides a numerical condition rating for applicable wall elements, which asset managers then can consider objectively relative to other influencing factors (such as the consequences of wall failure) to arrive at a recommended action.

Other influencing factors include the cultural and historical significance of the structure—an important aspect for the park program—and the reliability of the condition assessment data. The result is the selection of an appropriate action founded on a well-documented element condition and wall performance assessment, suitable for development of repair/replace work orders and associated cost estimates. The current wall assessment methodology meets the comprehensive WIP goals of identifying walls in need of maintenance, repair, or replacement; facilitating statistical assessments of wall elements throughout the entire WIP database; and providing a baseline for future wall assessments.

Initial Park Inventories

Field data collection, storage within the WIP database, and transfer to the NPS data management system began in April 2007. By February 2008, the inventory team had completed assessments on more than 3,500 retaining walls in 33 national parks, monuments, and recreation areas. This initial inventory, thought to encompass the majority of retaining wall structures within the NPS road system, serves as the basis for updated program developments included in the soon-to-be-released FHWA publication National Park Service Retaining Wall Inventory Program—Procedures Manual.

Aside from providing data on wall-specific deferred maintenance to the NPS data management system, the WIP database also can be queried to characterize and evaluate aspects of the entire NPS retaining wall asset—one that until now had been undefined. General findings to date include the following:

Wall functions. Of the six wall functions inventoried in the WIP—fill walls, cut walls, headwalls, switchback walls, bridge walls, and slope protection—approximately half represent outboard fill walls (downslope side of the roadway). If including culvert headwalls as a type of fill wall, then nearly 90 percent (based on WIP data) of all walls are designed and built to retain fill. Culvert headwalls supporting roadway assets comprise an overwhelmingly large percentage of the wall database. This leads to the question, as agency owners move to inventory and manage their culvert assets, where should culvert headwalls be included? The NPS inventory results show that culvert headwalls typically are small gravity structures in generally good condition. Including these structures in the inventory tends to bias and mask database performance trends for what could be considered the more traditional retaining walls, suggesting they are more appropriately assessed under culvert inventories. In comparison, cut walls comprise approximately 10 percent of the inventory, and a very small percentage of the walls are classified as slope protection, switchback walls, or bridge walls.

Wall types. Although the inventory team identified 17 distinct wall types, few dominate the database. Nearly all culvert headwalls, and 40 percent of all walls, are mortared stone masonry gravity structures. Dry-laid stone masonry walls comprise another 35 percent of the inventory. Most of these stone masonry structures were built in the first half of the 20th century. Of the 15 different wall types making up the remaining 25 percent, concrete gravity and concrete cantilever walls are relatively common. The inventory has only a few segmental block MSE walls and metal crib walls, and only one MSE wall with a geosynthetic wrapped face. "The distribution of wall types is probably indicative of the setting where and when the walls were constructed, and the relatively narrow timeframe during which most were built," says Keough. "Different owners and DOTs might find a completely different distribution."

Wall element ratings. As noted earlier, the number of wall elements rated varies with the different types of walls, generally ranging from 5-15 elements depending on the number of wall components and setting features. Nevertheless, when the team calculated the overall wall ratings, the maximum, mean, and minimum ratings remained generally consistent across the various wall types, indicating the WIP successfully quantifies wall condition within a reasonable band and with enough variation in scores that prioritization is possible.

Recommended actions. Thus far in the program, and for most wall types with significant populations (the number of inventoried walls in each type category), approximately one-third of the walls require some type of corrective action, mostly either maintenance or minor repairs to localized elements. These actions are relatively low cost and could be incorporated into a routine maintenance program. Less than 3 percent of the inventoried walls have recommendations to replace all or parts of the wall, suggesting that the asset as a whole is still in acceptable condition, and that a recurring maintenance program would go a long way toward keeping it that way.

Work order cost estimates. Any time an inspector recommends a maintenance, repair, or replacement action, he or she needs to prepare and submit a work order to the NPS data management system defining the work items and associated costs. Based on the NPS inventory, maintenance recommendations are most common and least expensive, averaging about $4,000 per wall. Recommendations to repair or replace localized wall elements are less common and have average costs ranging from $25,000 to $35,000. Total costs for wall replacement average about $150,000. The team estimates total deferred maintenance costs to date are approximately $18.5 million, with an estimated inventory replacement cost of nearly $407 million. This total equates to a program-wide facility condition index of 0.045, a relatively low index value, which further corroborates the belief that the total wall asset—all of NPS's retaining walls and other equipment as a whole—is in reasonably good health.

Future Efforts

The results of the program to date indicate that conducting a comprehensive wall inventory across all remaining NPS properties is not necessary to adequately characterize the condition of the overall asset and facilitate planning for annual catch-up on deferred maintenance activities.

Although specifically designed to serve the NPS data management system, the WIP methodology could prove useful for application among a broader national audience, as Federal, State, and local agencies look to tackle issues regarding their retaining wall assets. "This asset management system for retaining walls is both versatile and simple," says Daniel Alzamora, a geotechnical engineer with the FHWA Resource Center, "and it could be a cost-effective model for others to adopt."

Matthew J. DeMarco leads the Geotechnical Team for FHWA's Central Federal Lands Highway Division. Before joining FHWA in 2001, he managed private engineering consulting services and Federal research programs within the mining industry. He has a bachelor's degree in mining engineering from the Colorado School of Mines and a master's degree in geotechnical engineering from the University of Colorado.

Scott A. Anderson is the Geotechnical and Hydraulics team leader for the FHWA Resource Center. Before joining FHWA in 2002, he was an assistant professor of civil engineering at the University of Hawaii and a senior consulting engineer with URS Corporation. He has B.S. and M.S. degrees in geology from the University of Colorado and Colorado State University, and M.S. and Ph.D. degrees in civil engineering from the University of California, Berkeley.

Amit Armstrong manages the technology deployment program at FHWA's Western Federal Lands Highway Division in Vancouver, WA. He has been with FHWA for 7 years, coordinating deployment of new, innovative, emerging, and under-utilized technologies in design and construction of roads on Federal lands projects. He has more than 20 years of experience in numerical simulation and visualization of natural systems and is a licensed professional engineer. He received his doctorate in civil engineering from Texas Tech University.

For more information, contact Matthew DeMarco at 720-963-3520 or matthew.demarco@fhwa.dot.gov, Scott Anderson at 720-963-3244 or scott.anderson@fhwa.dot.gov, and Amit Armstrong at 360-619-7668 or amit.armstrong@fhwa.dot.gov.