Tools for Quantifying Risk in Bridge Management Systems
The Moving Ahead for Progress in the 21st Century Act (MAP-21) requires the development by each state of a
risk-based asset management plan and declares that “it is in the vital interest
of the United States … to use a data-driven, risk-based approach and
cost-effective strategy for systematic preventative maintenance, replacement,
and rehabilitation of highway bridges and tunnels to ensure safety and extended
service life.” Bridge Management Systems (BMS) such as American Association of State Highway and Transportation Officials (AASHTO)Ware Bridge Management
(BrM) are increasingly incorporating software capabilities to consider risk in
priority setting and life cycle cost analysis.
National Cooperative Highway Research Program (NCHRP) Project 20-07 / Task 378 is preparing a literature
review and synthesis of existing risk assessment methods that may be adaptable
for use with BMS software. The risks under consideration include earthquakes
and other types of earth movement, hurricanes and tornadoes, floods and scour,
fires, vehicular or vessel collisions, fatigue, and advanced deterioration.
These hazards have consequences that are beyond the normal deterioration and
functional deficiencies already assessed in bridge management systems. The
probability of each hazard, and the ability of each structure to resist the
hazard, is not always consistently and quantitatively assessed at the current
state of the practice.
The NCHRP Project 20-07 / Task 378 work is uncovering
significant gaps in the current state of the practice. Although data sources
exist for many of the relevant types of risk, many transportation agencies may
have difficulty in analyzing these data sets to objectively characterize the
likelihood and consequences of adverse event scenarios on the individual
bridges in an inventory using readily available data. Rather than asking each
state individually to perform research and analyze data to quantify risks,
there may be considerable value in developing national research-based
quantitative likelihood and consequence models that any agency can use without
additional research effort.
The objective of this research is to develop a set of tools
(mathematical formulas) to quantitatively estimate the likelihood (probability)
and consequences (cost in dollars) of transportation service disruption related
to specific types of risk, by analyzing existing data sets obtained from FEMA,
NOAA, USGS, FHWA, and state agencies. The research should be resourceful in
gathering relevant data for analysis, but the resulting models should be
capable of being applied in BMS by transportation agencies using just the data
items currently defined in the National Bridge Inventory and the AASHTO Manual
for Element Level Bridge Inspection. Such data would include for example location
(county, latitude/longitude, etc.), structure type, clearances and load
ratings, navigation features, and element composition and condition. If any new
data items are required from the inspection process (such as scour information
or fatigue details), the tools should include strategies to minimize the amount
of such data required.
Since risk models deal with fundamentally uncertain
phenomena, the tools should be accompanied by information on the degree of
uncertainty in estimates, to the extent that this can be developed. The usage
of these tools in BMS would primarily be as a means of consistently considering
asset-based risk in priority-setting for replacement and risk mitigation
projects. So broad coverage and consistency are more important than precision
in these estimates. Specific models of interest include:
- Probability and recovery cost of over-height
truck collisions as a function of bridge vertical clearance.
- Probability and recovery cost of bridge
overloads as a function of bridge load ratings.
- Probability and recovery cost of seismic and
earth movement events as a function of location and other relevant bridge
- Probability and recovery cost for hurricanes and
other weather events involving high winds, storm surge, and/or tsunamis.
- Probability and recovery cost of failure of
structural elements due to advanced deterioration and/or fatigue.
Proposers are asked to offer creative approaches to risk
quantification and to suggest additional hazards and data sources beyond what
is discussed here. The mathematical formulas developed in the research may be
demonstrated using spreadsheets or other types of calculations, but no software
development is expected under this project. Owner agencies will want to
incorporate the models into their existing bridge management systems using the
functionality of those systems.
Transportation agencies currently lack objective,
quantitative methods to assess risks for bridge management systems. The
research will fill in gaps identified in NCHRP Project 20-07 / Task 378 to provide
tools to objectively prioritize risk mitigation actions. The development of
these tools will enable many agencies to reduce the necessary effort on data
collection and analysis in order to implement risk management.
With these tools, agencies will be able to allocate
resources efficiently with bridge-level risk fully considered, and will have a
means of compiling and maintaining a detailed risk register. The public and
stakeholders will be in a better position to understand the trade-offs among
risk, funding, and other aspects of transportation system performance.
The National Bridge Inventory provides minimal opportunities
for risk assessment. AASHTO provides more relevant data items in its Element
Inspection Manual and in the AASHTOWare Bridge Management software. However,
many of the available items are largely unused due to lack of guidance on how
to use them in asset management analysis and decision making. NCHRP Project
20-07 / Task 378 documented methods for incorporating risk into BMS, but some
of these methods require considerable additional analysis of national data
A very comprehensive vulnerability assessment process is
documented in a series of manuals in New York. Florida DOT has completed
research on estimating likelihood of a variety of natural and man-made hazards,
and has demonstrated one method of incorporating this information in a bridge
management system. Minnesota’s Bridge Replacement and Improvement System
demonstrates a process where risk assessment is used for prioritizing projects
for inclusion in the STIP. NCHRP Report 590 documents some alternative methods
for combining risk with other performance measures in a multi-objective bridge
management analysis. Transportation Asset Management Plans in all 50 states are
starting to develop concepts for incorporating risk into the highest levels of decision making in each agency.
The emerging field of geotechnical asset management, where
risk is a primary performance concern, is developing methods for routinely
assessing the geological and hydraulic risks to each section of road, and
quantifying the effects of slopes, embankments, retaining walls, culverts, and
other constructed facilities in resisting these hazards. The methods often
adopt a level of service perspective, where the “service” is the reduction of
event likelihood or the increase in road section resilience. This experience
may provide some useful ideas for bridge management.
It is envisioned that the research will involve at least the
Augment and update, as needed, the NCHRP Project 20-07 /
Task 378 literature review on risk assessment and risk management. Of
particular interest are data sets available from national, state, and
international sources that can help in quantifying the likelihood and
consequences of hazard scenarios. Contact researchers and practitioners in a
selected group of state DOTS to identify existing research studies which might
have helpful data sources. Identify potential sources of anecdotal records of
extreme events that may fill in gaps where systematic automated data sets are
Document an analysis plan as a set of models and
alternative approaches to each type of hazard, which the researchers believe
may be capable of being developed using the available data sets. Document the
hazard scenarios to be considered. Identify potential problems and limitations
of different approaches to quantifying likelihood and consequence of
transportation service disruption. This analysis should especially consider the
extent of national coverage, the ability to reflect state-specific and local
conditions, the ability to work with data already found in bridge management
systems, and the potential for future validation and improvement of models.
After approval of the analysis plan, carry out the plan
by gathering data sets, performing statistical data analysis and model
estimation, performing validation of statistical models, and evaluating the coverage,
quality, and usefulness of the models. Multiple alternative approaches are
encouraged in order to find the approaches that provide the most useful insight
for priority setting in risk management applications.
Prepare a Final Report (draft and final) to document the
research. The Final Report should contain a User Guide chapter capable of being
extracted and implemented by transportation agencies in conjunction with the
products of NCHRP Project 20-07 / Task 378.
This research is envisioned to feed directly into existing efforts to develop multi-objective optimization models in bridge management systems. It will provide default models and methods each agency can use as it customizes BMS models for its own use. Continued FHWA and AASHTO support for bridge management systems, and in general for asset management and performance management, will help to ensure successful and widespread implementation.
|Sponsoring Committee:||AKT50, Bridge and Structures Management
|Research Period:||24 - 36 months|
|RNS Developer:||Paul D Thompson|
|Source Info:||AHD35, Bridge Management|
|Index Terms:||Asset management, Risk management, Bridge management systems, Data files, State of the practice, Geological events, Atmospheric phenomena, Metrics (Quantitative assessment), |
Data and Information Technology
Maintenance and Preservation
Planning and Forecasting
Bridges and other structures