From Our Partners – A Philosophy for Designing 22nd-Century Infrastructure

Posted: October 27, 2015 at 11:00 am, Last Updated: October 29, 2015 at 9:44 am

Print Friendly, PDF & Email
Steven D. Hart, Ph.D., P.E.

For the past 30 years as an engineering student, practitioner, and educator I have been intimately involved with design philosophies and approaches.  With my focus on infrastructure issues over the past eight years, I have come to the conclusion that our current design philosophies and approaches do not serve us well when it comes to infrastructure because the assumptions and perspectives implicit in the approaches are not consistent with the nature of infrastructure.  Just as much of our current infrastructure was designed and built in the early and mid-20th century, today we are designing and building infrastructure that will endure into the 22nd century.  If we are to do this well, it will require a new design philosophy consistent with the realities of infrastructure and the needs of the society it serves.

Our current philosophy of design uses a standardized approach which has the presumption of a design life and a binary perspective on adverse events.  The standardized approach tells a designer when faced with a certain situation, “apply the loads from a table, perform the required calculations, and satisfy the given limit states.”  Most of these calculations are based on expected probabilities within the assumed design life of the structure, which varies by code and country, but is typically about 50 years.  For example, the Australian Building Codes Board lists building design life as 15 years for short, 50 years for normal, and 100 years for long;[1] the EUROCODE gives the design life for temporary structures as 10 years, agricultural buildings as 30 years, common structures as 50 years, and monumental structures as 100 years;[2] and the UFC 1-200-01, General Building Requirements, describes as permanent facilities expected to serve for 25 years or more.[3]  Basically, we expect our structures to wear out and be torn down.

Within all of these codes, the fundamental design approach is to ensure that the strength of the individual members and the facility as a whole is sufficient to carry the required loads.  Required loads, whether they are the occupancy load of a building, winds, earthquakes, or floods, are based on the probability of occurrence within the expected life of a structure.  For example, when the National Flood Insurance program was established in the 1960s, the flood level with a one percent annual exceedance probability, commonly called the 100 year flood, was selected as the base flood event because it “was thought to be a fair balance between protecting the public and overly stringent regulation.[4]”  As long as the designers ensure that the facility can withstand the code prescribed loads, then they have done their jobs.

So why doesn’t this approach work for infrastructure?  Simply put, infrastructure is different.  “How,” you ask.  Well, consider these observations:

Infrastructure problems are social problems.  Although they have strong technical components, infrastructure problems are, at their root, social problems because infrastructures deliver goods and services that both people individually and society collectively need to live and function.  Without electricity there is no water, without petroleum there is no food, and without food and water, there is no society.  Furthermore, the norms and values of the society shape what is acceptable and necessary in a solution.  If I were to state that I could solve all of our city’s energy needs for the next 100 years by putting a nuclear reactor in each elementary school parking lot and painting all of the city’s landmarks bright purple, people would refuse because the solution does not provide the kind of city people would want to live in, even if the energy were free.

Each infrastructure problem requires a unique solution.  Because each society served is different in location, demand, need, and capacity to withstand service disruptions, each infrastructure problem is essentially unique.  The solution for the drinking water supply in New York City, population 8 million, cannot be applied to Goshen, VA, population 360.[5]  Where a farm family in central Kansas expects to survive for two weeks without power during a December ice storm and is fully capable and prepared to do so, a family dwelling in a 37th floor apartment is hard pressed to survive for six hours without electricity.  One family is willing to pay much more than the other for the rapid resolution of a blackout and needs that rapid resolution.

Infrastructure does not have a design life.  Consider monumental structures such as the Pentagon (completed 1943), the Empire State Building (1931), Faneuil Hall (1742), and the Tower of London (1078).  When will we tear these down? Never.  Now consider our interstate highways, major airports, water treatment plants, or Class 1 railroads and ask the same question.  When we build major elements of infrastructure, history has shown that we do not tear them down—we patch, we fill cracks and we may replace major components, but we expect them to be around for a long, long time.

Infrastructure use exceeds our vision.  Infrastructure routinely outlives its designers and builders, and its future uses nearly always exceed those envisioned during planning.  When the Brooklyn Bridge was completed in 1883 it carried horse drawn carriages, rail traffic, and pedestrians.  Later horseless carriages replaced horse kind and in the 1950s, the bridge was reconfigured to carry six lanes of automobile traffic.  Later, during both blackouts and terrorists attack, the bridge was full of pedestrians as people were forced to walk off Manhattan Island.  The uses of the late 20th and early 21st centuries could not have been anticipated in the original design.  Who knows what the bridge will carry in the next 100 years.

Infrastructure constrains the future.  Once built, a significant piece of infrastructure becomes a design constrain for future work.  When the Panama Canal was designed in the first decade of the 20th century, the size of the locks was set at 1050 feet long and 110 feet wide,[6] dimensions which constrained the size of United States Navy ships through the 1960s when the first super-carriers were built.

Adverse events and failure are not binary. In design codes, maximum loads or event return intervals are stated in a table, for example, 50 pounds per square foot for an office live load or the 1 in 2,500 year earthquake.  If the proposed structure can carry the load, then the design is ”good,” but if the actual loads exceed the design loads, then the structure ”fails.”  Actual load events, rather than having only one option, occur in a full spectrum from very small, to the design load, to very large and the failures that occur will vary in severity based on the loads.  For example, the 1 in 3,000 year earthquake will produce a very different level of building damage than the 1 in 10,000 year earthquake, yet the design code has only one term to describe both:  failure.

So, if current design approaches are ill-suited to designing 22nd century infrastructure, what elements should be included in an infrastructure design philosophy?

  1. Plan for perpetuity. Instead of a cradle-to-grave approach (construction, occupancy and maintenance, demolition, disposal) or a cradle-to-cradle approach (construction, occupancy and maintenance, deconstruction, recycling, construction) we should accept that major infrastructure elements will never be torn down and plan for perpetuity with a life cycle as shown in Figure 1.  Planning for perpetuity leads to building in right from the beginning the ability to remodel, renovate, and repurpose the facility.
  2. Hart Figure 1Figure 1 Perpetual Building Life Cycle
  3. Design for success and control failure. When planning for perpetuity, which is a rather long time, we are forced to accept that we will not be able to withstand all adverse events without damage. Figure 2 shows the yearly and cumulative probabilities of flooding events over three periods in a building life cycle.  For a building with a 100 year design life, there is a 63% cumulative probability of seeing the 100 year flood, which is a high enough probability that is should be included in the design, but only an 18% chance of seeing a 500 year event, a low enough risk to be reasonably accepted.  However, if the building is expected to last 200 years or more, the 100 year event becomes a near certainty while the chances of a 500 year event are 1 in 3, which is really too high to ignore but might not be quite high enough to cause it to become the basis for design, especially design on a budget.
    Hart Figure 2
    Figure 2  Flood Event Probabilities

    Given that adverse events, and the damage they cause, are not binary, we should respond with three loads and performance levels.  The design load caused by moderately severe but high probability events should result in only cosmetic damage to the facility. The repair load, which is generated by more severe but less likely events, would damage the structure, but that damage would occur in a controlled manner that facilitates repair and restoration.  Finally, the loss load, which corresponds to the risks of events we are willing to accept and do nothing about, is the event which we expect will destroy the facility where the remediation is demolition and new construction.  With this design approach we can say that our facility is “designed for the 100 year flood, resilient in the 500 year flood.”[7]

  4. Right is good, adaptable is better. As designers, we like to get things “right” and satisfy clients, society, and the outstanding projects awards committees from our professional organizations.  Given that our creations will outlast us and still be in use by our great, great grandchildren, how do we get it ”right” for them?  We can do this by designing facilities that can be easily expanded, reconfigured, and repurposed and by minimizing constraints on future use and development.  Simply put, our structures must be adaptable so that our great, great grandchildren can tailor them to their need and tell their children, “You know, those designers back in the 2010’s were really smart to set this up for us like this.”  In doing so, we make uncertainty about the future irrelevant—our successors will solve those problems because we enable them to do so.
  5. Satisfy society, not codes. Acknowledging the nature of infrastructure problems and applying this design philosophy means that satisfying design codes is a necessary part of design, but not sufficient.  Our focus must be on satisfying the needs of the society served based on its unique characteristics and the unique characteristics of each project.  This means that our highly standardized designs of today must evolve into unique designs for the future.

Implementing this philosophy of infrastructure design will certainly be challenging.  It will require a new level of cooperation and communication between owners, designers, builders, code writers, and building officials.  It will require leaders who are comfortable with ambiguity, capable of accepting and managing risks, and deeply committed to simultaneously serving their own firms, their clients, and society at large.  Challenging, certainly, but better than the alternative of another century of news reports about our aging and crumbling infrastructure.

Acknowledgement

The author gratefully acknowledges the assistance of Dr. Led Klosky of the West Point Department of Civil and Mechanical Engineering and Dr. Paul F. Mlakar, Engineer Emeritus of the US Army Engineer Research and Development center for reviewing this paper and making substantive recommendation for improvements.

About the Author

Dr. Steven D. Hart, P.E. is an adjunct professor in the Department of Civil and Environmental Engineering at the Virginia Military Institute, the Chief Engineer of Hart Engineering, LLC, and an aspiring gentleman farmer at Hart Burn Farm.  His research areas of interest include infrastructure engineering, infrastructure education, infrastructure resilience and security, and grass-based sustainable agriculture.


[1] Durability in Buildings,  Australian Building Codes Board, (2006): 5, available at http://www.abcb.gov.au/education-events-resources/publications/~/media/Files/Download%20Documents/Education%20and%20Training/Handbooks/2006_durability_in_buildings.ashx.

[2] European Committee for Standardization, EN1990 EUROCODE Basis of Structural Design (Brussels; European Union, 2002), available at https://law.resource.org/pub/eu/eurocode/en.1990.2002.pdf.

[3] “UFC 1-200-01, General Building Requirements,” United States Department of Defense, Aug. 1, 2015, available at http://www.wbdg.org/ccb/DOD/UFC/ufc_1_200_01.pdf.

[4]Robert H. Holmes, Jr., “The 100-Year Flood—It’s All About Chance,”U.S. Geological Survey, Last Updated July 30, 2015, available at http://water.usgs.gov/edu/100yearflood-basic.html.

[5] Google Public Data from US Census Bureau, Updated July 24, 2015, accessed  Oct. 12, 2015.

[6] Jeff Brown, “Between Two Oceans: The Panama Canal,” Civil Engineering 84, no. 7 (July/Aug. 2014), 42-45.

[7] This phrase is not mine.  I heard it on a conference call in 2009 but do not know who said it so I cannot give the correct attribution.

Write to the Editors at ciprpt@gmu.edu