Back to Agenda
FHWA Unknown Foundations Summit
Where Do We Go From Here?—Panel Discussion
Larry O'Donnell, moderator, FHWA
Keith Ramsey—Texas DOT
Jean Louis Briaud
Roles and Responsibilities
What should be role and responsibility of FHWA?
Jorge Pagan: develop guidance for bridge owners and help manage with local owners. Assist states make decisions on issue.
Keith Ramsey: act as clearinghouse of information for states, not just guidelines.
Ineichen: Mandate it. Come up with variance, tell permit people to back off. Provide research and development.
Sabol: Good role as clearinghouse, champion success stories to share among states.
Briaud: Leadership and direction. Get back to reasonable budgets and free choice of where money goes.
What should be role and responsibility of bridge owners?
Ramsey: Texas DOT does inspections. Have standardized format. State owners are responsible to traveling public. Problems of UF falls on our shoulders. Reduced by half and are still doing so. Remaining UF bridges were taken over from other owners. There are no plans available for a lot of bridges, therefore lot of UF bridges. Lots are on low ADT roads, 6800 bridges of 8,000. Want to see prioritization. Can't determine 100% of all structures.
Beard: Bridge owners have ultimate responsibility for public, need to make sure they're safe.
Briaud: Must do whatever we can, but can't get to zero risk. What is acceptable risk?
Role & responsibility of industry?
Stu: Need highly trained folks. Need technology to be accessible to lots of engineers. We develop model and give it to client for them to use. Need to do that with NDT. Accessibility is always going to be an issue.
Role & responsibility of academia?
Briaud: I have to develop new knowledge and then teach it.
Sabol: First is to come to these events. For industry, it would be wonderful to know what internal protocols they use to approach these projects. For academics, we have been slow to evolve in civil engineering, we tend to teach how to design new stuff, not as much recognizing and reusing existing infrastructure. We need to evaluate current existing infrastructure.
O'Donnell: Audience comments?
Q: FWHA should take major initiative, and then come up with program elements that include geotechnical, geophysics under R&D. Take lead to establish GIS system for UF. Must be full funded.
Q: Accessibility of products. New technologies will come out and will have magic box. It comes with time, but industry has got to grow and develop. Better new technologies.
Q: Khamis: This meeting failed to invite students to come and listen. Need to encourage students, maybe give ideas for their research.
Q: Bridge owners have to balance lot of responsibility and lots of effort with very little money and limited resource. For FHWA and research, SAFETEA-LU. Congress earmarked every dollar for research and none of that money is for hydraulics, geotech, etc. Until 2009 we have to work under this. The option is to support research through NCHRP or from FHWA coordinating multiple state study.
Q: Need spend money on studying process instead of relying on age and experience. Need to itemize and spend money to ensure that all of our knowledge and wisdom is not wasted. We've been harassed by 21 major hurricanes in Florida last year alone. We learned something new every time. Takes lot of time and effort to bring knowledge together.
Management Strategies and Guidance
What is your perception of the amount of emphasis that is being focused in this area?
Ineichen: Not much emphasis outside of our own dept.
Pagan: Emphasis is at a 2, 1-5 scale. We recognize this a problem, but we have deferred different action. Emphasis has gone up to a 3 in this past year.
Ramsey: UF are important but not every UF has same risk. Lots of tools are available that state DOTs are using. Lot of info could be used from this to help manage the situation. We need to prioritize and address those first.
Briaud: In scour critical situation with UF how do you categorize.
Pagan: For prioritization needed, be more important take action on bridge that is scour critical. But also need take action on UF bridges.
Briaud: No unknown foundation bridge can be scour critical?
Pagan: Tough decision to make, can't decide without knowing foundation depth. Without knowing …
Beard: UF vs. scour critical. We classify them as unknown. Can't code if don't have foundation. Did test for foundation. Once known, then can be coded.
How improve current strategies?
Stu: Need performance goals. Have decisions to make, lots of technologies available. Need to assess risk if don't make that choice. Risk not just economic, but probability that risk will occur. If never had a problem with scour on an old bridge, expensive scour test probably not needed. Need to decide what want out of this. Need look at performance for all bridges.
Sabol: Two things that could help. FHWA needs to recommend improved mining of data. Second, devise method that reduces new unknowns. Don't know how to do it, but important and needs to happen.
Briaud: Need to think every time there is a number there could be an error. Need to think in terms of standard deviation.
What additional management strategies are needed?
Pagan: Need revise technical advisory to go over many issues for overall scour program, and for UF. Should be more proactive.
Beard: More guidance on training for item 113. Hard to code bridges. Inspectors used to use 1 sheet of paper. Inspectors get some training courses; we should be involved in that training. Help to have more guidance for typical bridges. In examples, shows illustrations, but no example for pile bents. Training would help a lot with coding.
Q: Use logical numbering system. Current system doesn't make a lot of sense. Also save time doing numbering manually.
Pagan: Internally, we recognize more can be done and hope to improve logic.
Q: Colo DOT: We rate based on load-carrying capacity. If follow that load path, xx. For substructure we don't have rating. No Q/A for a rating check. Not another expert checking on the process. Need hydraulic structure into action. Maybe we should rate with survival, like the traffic load. Water pressure is also a load. If we can determine performance requirement and have to rate it when we finish, that will help greatly. Our plan and details is all 2-D. We might have missed something important. Need national standard. Have rate survivability with regard to scour.
Q: Lot of States that are looking at scour critical not just flooding. Alabama has invested lot of money for research.
Q: Should have some risk analysis.
Q: In Indiana, we are providing information that includes table. Table covers friction, if expecting negative drag, ultimate load.
Q: We're mostly structural engineers, but hydraulics engineers need make more than sand model.
Pagan: We made extra effort to get out info on soils. We attempted best effort to provide guidance on soil and rock. There are recommendations on scour for sand bed but also sections on rock as well.
How can current risk management tools be improved?
Stu: Lots of states have management tools and good strategies out there, but rarely documented. We interview folks. States need to document this. Some states might categorize consequence for ADT, but not risk. Not lot states look at risk. Need decide on probability of structural failures. Don't know what risk is for UF bridges.
Ramsey: Advocate there be more effort be put into more risk management tool criteria to put into to determine system wide approach. Existing tools need improvement. Need risk state DOT owners can live with.
Florida: Experience is major tool. One of our criteria is what happened to bridge in previous hurricanes. If don't see major change, put bridge in low-risk category.
Briaud: First calculate risk, then consider what is acceptable. Need data; experience is data. You can calculate current risk you are taking. Need to decide if risk is too high or too low. If choose risk, what is economic impact? Need to discuss this.
Beard: Coding bridges is risk management. Taking lots of risk categories when coding bridge. It is going through process when coding bridge. RM is good tool to put factors together. For initial screening. Countermeasures, could get into trouble if don't know what's in field. Need to know what's in field before design countermeasures. RM is good first tool to go through process.
Sabol: Need to be careful using phrase "unknown foundation." Might want to explain to layperson what UF is. Don't want Congress to get involved and mandate, perhaps incorrectly, a "solution" to the problem.
Q: Bridge engineers are risk averse, more concerned about politics than safety. One issue is that of acceptable risk. Brought to AASHTO in past. Engineers don't have enough knowledge to know what is acceptable risk. We asked for specifications, they said we don't know. AASHTO guidelines aren't written by bridge engineers. Doesn't matter how good RM tools are if decisionmakers don't know what this all means. A common hazard way of looking at all hazards. Need to take all extreme events and create framework. Create framework is difficult. But take all factors and put them on a common hazard way the evaluates risk.
Briaud: There are numbers, but we don't know where they are coming from. Maybe we should change that.
Q: Kentucky has several different ratings classes. Political consequences might be more palatable to decision makers. Some ideas of consequences and military response. One comment is that risk idea, what I've been envisioning is that FHWA create a procedure. NDE would be one tool of investigation process.
Stu: I think guidelines developing for NCHRP are what you're talking about. But it doesn't get into how do it. Others will tell you. I would expect states to be risk averse. Policy decision that states should make.
Q: Good to hear about high tech models, but what do we do now with problems we have now. Want to see directive right now. I need answer right now. Don't leave subject floating. Need solution now. We need an answer now. Second, statistics should be signed by company doing testing. At end of each project, someone comes up with folder containing all information.
Beard: We get report from consultants, signed and sealed. Contains information. Sent to our bridge maintenance section, entered into main database. An inspector can use this info next time they go out.
What additional research is needed?
Briaud: If evaluate loss at $100 million per year, what percent invest in research. If we can't decrease cost by 10 percent doing something wrong.
Stu: For structural integrity want to see design standards. Need to look at failures and what caused them.
Pagan: Integrity is important, needs be looked at.
Sabol: Integrity, I'm in favor of pulling piles to see what is actually there and if it matches what was expected. Capturing knowledge and developing knowledge-based expert systems. Eliminating guess work out of interpreting NDT data. Clean up all the NDT pictures. "Need some tools now." Get something down on paper. Figure out how inspire next generation of thinkers to predict in advance what the next big problem facing the bridge infrastructure may be. Get some people to think about what next set of problems instead of just wait for to arrive.
Pagan: Come up with better ways to allow normal people to interpret what's coming out of geophysics.
Ineichen: Like to see more scour research done on variety of soils. Geotech is failing it but it's basically going to risk management. More research done on that aspect. Want to reduce pile movement. Streamline collection processes. If safe effective way to find out pile capacity is.
Hawaii: No directives, leave it to the states. Standards are out of whack, retaining walls are huge. Up to the states on how to do it. States are different, not all things work for everyone.
Q: No timeframe for guidelines to be delivered.
Pagan: It's our role to develop policies. Policy is bridges over waterways to be evaluated for scour. Also put together guidelines. Up to bridge owner to follow guidance.
Briaud: We have ability to test samples. But if use 1 method get large numbers and don't have proof if is accurate. Need to undo our scour evaluations on known bridges. Lot of bridges getting classed U because no body wants to put their stamp on it.
Technology Deployment & Development
What should be done to better deploy current technologies?
Pagan: Idea of creating means be good way to promote on nationwide basis.
Q: If our state had more guidance for outcomes and goals, we could allocate money easier if management could see goals. With no goals, have hard time allocating money. FHWA should provide guidance.
Q: Use a monitor device to measure actual scour with calculated scour. Then project to unknown foundations. The area that we have the devices can be check real quick. It's low cost but we need funding.
Q: Problem is could be $40,000 or $50,000 to monitor one pier. More expensive than you think.
Q: We are repeating some work. Need GPR to get representation of ground. Use for various analyses: hydraulic, seismic, geotech. Not enough coordinated work. Need 3D profile under bridge. Put effort together to eliminate overlap.
Q: Geotech engineering is not pure science, it's art. Geotech engineer decides how to use information. Need more skilled people to use equipment. Need more trained people. Geotech problem is a xxx problem. One mistake can create problems.
What type of training is needed?
What might be a reasonable duration for training courses?
Ramsey: Needs to be general subject matter. Standardization so people are on same page. Can be used for disciplines as well.
Briaud: Familiar with Web courses? Webinars. Be easier to attend.
O'Donnell: 4-day training course too much?
Beard: Get people together, go over topics, get all people on same page. Week-long course is reasonable. Haven't talked much about countermeasures, good to see what other states are doing. Could spread this knowledge around the country. UF course would be great.
Q: Schedule training course at site where are performing investigations. Participants could go out to site and witness investigation hands-on.
Q: Khamis: who would do training? Should FHWA develop course and have someone else do training? Companies tend to be biased and want to sell their products.
Pagan: Effective course on coding, NDE, etc., FHWA and DOTs should be involved so all can share experience.
Sabol: Who should train? If have 1 industry person, could be bias, but if have 2 industry competitors could be more unbiased.
Q: Khan: One week has been a problem. Not everyone can do 1 week. Keep as short as possible, 2-day max. Keep costs down, keep attendees down.
Q: Colorado could use a video course to keep in library. Take them at leisure and work through them. We have interest in rebuilding infrastructure.
Q: Maybe DOTs do phased approach. Have a few people take training. Data could be processed by consultants, but DOT people go over data.
Back to Agenda