U.S. Department of Transportation
Federal Highway Administration
1200 New Jersey Avenue, SE
Washington, DC 20590

Skip to content U.S. Department of Transportation/Federal Highway AdministrationU.S. Department of Transportation/Federal Highway Administration

Federal Highway Administration Research and Technology
Coordinating, Developing, and Delivering Highway Transportation Innovations

This report is an archived publication and may contain dated technical, contact, and link information

Publication Number: FHWA-RD-03-089
Date: September 2005

Dynamic Bridge Substructure Evaluation and Monitoring

Previous | Table of Contents | Next

Chapter 8. Integration of Dynamic Testing Results to Bridge Management Systems


A bridge management system must have a comprehensive bridge inventory that contains the number, type, size, and condition of each of the elements, the cost of maintenance and repair activities, and predictions for the future bridge conditions. Commonly used bridge management systems such as Pontis™ rely on subjective visual ratings to determine bridge element conditions. Recent research suggests that integrating NDE with visual ratings provide more consistent and richer bridge condition data.

Exploration of dynamic bridge substructure evaluation and monitoring systems has shown that bridge foundation vertical stiffness is an appropriate indicator for the bridge condition evaluation, and it can be used to support BMSs in three ways:

It is important to remember that a BMS is not a complete repository of bridge inspection-related data; however, it is valuable to think of a BMS as a framework that can be used to obtain the results. Hearn and Shim propose integrating NDE data with visual condition ratings to form an integrated condition state.(189) A hypothetical integrated condition state for a reinforced concrete bridge substructure is shown in table 17, which demonstrates the potential role of dynamic testing results in BMSs for assessments of bridge safety, scouring, and condition.

Table 17. Hypothetical integrated condition state for reinforced concrete substructure.

Item/State 1
Change in vertical stiffness <10% 10% to 30% 30% to 50% >50%
Foundation scour Not present Some Significant
Foundation movement Not present Some Significant
Corrosion No activity Possible corrosion Corrosion activity Loss of rebar section
Fracture No activity Some cracking Cracking and delamination Spalls
Action No action Observe Post load limit, reinforce, replace Close bridge, rehabilitation, replace

The Role of NDE in a BMS

Because a BMS relies on bridge condition ratings to develop optimal decisions for repair, maintenance, and rehabilitation, NDE results can be used for BMS in two strategies. The first is to modify a BMS to account for the additional information from NDE data. The second is to use NDE data to either determine or modify the bridge condition rating before it is input into a BMS. Using either strategy requires careful consideration of the quality and reliability of NDE results, and the different types of information that NDE and visual ratings provide. In general, NDE data require interpretation before becoming information that can be used to make decisions.(145)

Hearn and Shim have developed a strategy for integrating NDE data into or with condition ratings instead of modifying a BMS to account for additional information from NDE data.(189) By reducing the reliance on visual inspection, the integrated condition ratings reflect the presence of aggressive agents, the stage in the deterioration process, and the existence of damage. The proposed condition ratings are integrated with a focus on the condition states that are mutually exclusive, detectable, defined by multiple attributes, correspond to maintenance repair actions, and indicative of severity but not extent of damage. Table 18 shows an example of the integrated condition states approach for a reinforced concrete element.

Table 18. Integrated condition states for a reinforced concrete element, modified from Hearn and Shim.(189)

Attributes Exposure/condition state No ingress of chloride (Cl) ions Cl ion ingress, concentration below threshold Cl ion ingress, concentration at threshold
Corrosion activity level No corrosion activity Possible corrosion activity Corrosion activity
Rebar damage No loss in rebar area Loss of rebar section
Concrete damage Delaminations and minor spalls Large spalls
NDE Measure Electrical resistance High Low
Specific ion probe Low Cl High Cl
Corrosion current Low High
Impact echo, radar, sounding No damage Damage

Similarly, Hadavi suggests the use of NDE results to quantitatively measure condition states, which can be compared with a standard that uniquely determines the repair or maintenance strategy.(127)

The Role of Dynamic Testing Results in a BMS

Dynamic bridge substructure evaluation and monitoring provide bulk properties of the bridge substructure in terms of stiffness or flexibility, and they measure downward natural frequency shifts associated with damage.(4) The bridge foundation vertical stiffness could be used in the future at an aggregate level for monitoring the changes of the bridge foundation conditions such as stiffness or flexibility versus a baseline signature. Knowledge of the bridge foundation type and depth to a large extent may be obtained with surface and borehole NDE methods.(27)

Based on the identification of the bridge foundation type and depth and evaluation of the bridge foundation condition as sound or damaged with reduced capacity, the inventory of the bridges, particularly of those with unknown foundations, can be completed to assist in better decisions on overall bridge management strategy. Currently, Pontis does not differentiate between pile caps and footings, but it does distinguish pile caps and piles as well as footings and piles.

Changes in bridge foundation vertical stiffness from the baseline condition indicating either substructure damage or loss of bearing capacity, or both, should be an effective indicator for repairs or other actions, such as the posting of weight restrictions.(189) Experience with the limited set of bridges in this project and experience gained on drilled shaft foundations suggest that when the changes in bridge foundation vertical stiffness are greater than the threshold values to be determined in the future (hypothetical data shown in table 19), the appropriate actions should be taken. The suggested hypothetical threshold values are based on limited research data, and further investigation is needed to determine their appropriateness. Furthermore, research is needed on the HHT frequency shifts that indicate damage. The suggested threshold values most likely are conservative. The use of threshold values is applicable to both long-term monitoring and short-term evaluation after an event such as an earthquake, barge collision, or huge flood. Consequently, catastrophic failure could be avoided. Prevention of catastrophic failure means saving lives, saving time by avoiding detours and reconstruction, and preserving the existing investment in the bridges.

Table 19. Hypothetical thresholds for action.

Foundation State and Change in Bridge Foundation Vertical Stiffness from Baseline Action
Sound < 10% None
Questionable 10% to 20% Regular observation/continuous monitoring
Poor/Damage > 20% Action required ranging from reinforcement/repair to immediate action such as closing bridge depending on the severity of the vertical foundation stiffness reduction

Integration of Dynamic Testing Results into a BMS

Before dynamic bridge substructure evaluations can be widely adopted, additional experimentation and evaluation are required. Several practical issues also must be addressed and a data-collection strategy must be developed.


Additional experimentation is required to further develop the correlation between NDE attributes and bridge substructure conditions and explore field data quality. The NDE attributes are not necessarily direct measures of the bridge substructure conditions. Instead, the NDE attributes are simply indicators of performance of the tested structures.(190) The limited experimental results of this research suggest that the use of bridge foundation vertical stiffness and possibly the HHT method to identify frequency shifts are appropriate measures of performance of the bridge substructure; additional experimentation is required to confirm the relationships among bridge foundation vertical stiffness, HHT spectrum results, and bridge substructure load-carrying capacity, element condition, and remaining life. Because no other objective and quantitative measures of bridge substructure conditions exist, expert judgment is required. The HHT method may offer more accurate identification of weakened and damaged bridge foundations than the stiffness approach if the HHT method can be implemented on a comparative basis without having a sound bridge signature for comparison.

The limited sample of bridges used in this project did not permit any evaluation of field data quality. Hearn and Shim point out the need to explore accuracy and variability (repeatability) when using NDE.(189) Because bridge foundation vertical stiffness will not be monitored continuously, change must be distinguished from inherent variability. Recognizing that a change in bridge foundation vertical stiffness is a realization of a random variable, one may assume that different probability density functions are applicable to damaged and undamaged substructures, as shown in figure 161.

View alternative text


Figure 161. Graph. Hypothetical probability density functions for percent change in vertical stiffness.

Evaluation of NDE Methods

Cost/benefit analysis of an NDE method requires understanding the value of information obtained from the NDE method and the cost of the NDE data collection. This also involves developing strategies to select an NDE method to identify vulnerable bridges. Other important issues are the role of complementary versus competing data and opportunities for sensor fusion. One technique for evaluation of an NDE method is to use an influence diagram for the expected life cycle cost for the bridge with and without the NDE data.(191) While this is not easy, it does provide some important insights into the value of an NDE. Prine identifies the need to address several practical issues:(192)

Developing Data-Collection Strategy

Federal mandates require visual inspection of all bridges on the Federal-aid system every 2 years. Because NDE is not mandated, a recommended data-collection frequency must be determined. This provides an opportunity to design a data-collection strategy that reflects the need for NDE data. The strategy will identify the following information:


Dynamic bridge substructure evaluation and monitoring provide some important opportunities for improving bridge management systems and enhancing the current state of the art in bridge evaluation and monitoring. Specifically, measures of dynamic bridge foundation vertical stiffness or HHT results, or both, that identify downward frequency shifts indicating damage show promise in the following applications:

In conclusion, recent research on the role of NDE in BMSs suggests the desirability of integrating dynamic testing results, including HHT results, with visual ratings data.


Previous | Table of Contents | Next