Did the bridges in the sample have comparable span and total lengths? One would assume new bridges display a strong survivorship bias towards increasingly challenging sites.
I would have found the planning phase estimates to be more useful if they were devoted to the engineering/environmental related planning. I can't say for other readers, but when I think about how long a project is going to take I'm focused on the professional engineering planning (along with the construction period.) But using an announcement by politicians or bureaucrats at an agency as the start time isn't quite as clear cut as when the major professional contracts are let. When the civil engineers, and structural engineers are chosen and their clocks start, that could be a better start point. That would include enviro studies and surveys, but not include years of political fighting or zoning fighting or land purchase machinations. In sum, either when a competition is concluded and the designers are chosen or when a direct contract is established with the lead structural engineers, this approach would help limit what is otherwise a very open-ended idea of "planning" that leads to some of your listed projects going on for decades or a half century.
This approach would help avoid the future hypothetical example of the Strait of Messina bridge, which has been proposed for literally centuries, and which has had contemporaneous proposals since about 35 years ago. And even though some recent efforts have led to some engineering contracts, the reality is that the "planning" phase of the project hasn't started by my definition and only would do so when all the legal challenges are resolved and work is underway to do the detailed *design work* that leads to actual construction.
When people say "gee, it takes way to long to build anything", they're using a language of time, but I suspect really talking about money: "gee, we seem to be spending a whole lot of time and money before we actually put anything in the ground."
To that end, I'm not sure the planning metric is completely valid. "Let's take some extra time on the geology" is different from "let's spend millions saving some snail and moving 6 Swainson's hawk nests before we do anything." Technically they're both "planning" but there is a qualitative difference.
I'm not saying this to be critical of the latter behavior. Wealthier societies can worry more about snails and hawks; poor societies just need the stupid bridge built now. But I do think it makes your "planning" comparison somewhat less valid.
I didn't see the new Gordie Howe International Bridge between Detroit, Michigan USA and Windsor, Ontario CANADA. I believe it was scheduled to open for traffic sometime this month (perhaps it's already open). Do you have planning/construction times on this bridge?
Great article. I’m an environmental planner at a civil engineering firm. People sneer about birds’ nests and obscure plants but on most projects I’ve worked on the real resource sinks are historic/cultural resources (defined in an absurdly broad fashion IMO) and public engagement. The latter can trigger significant design changes, which of course cost a lot of time and money. Politics can also be tough for replacements that haven’t actually collapsed, especially when funds are limited. It’s a lot more fun for officials to announce new projects.
"Historical/cultural" in Canada means any land where aboriginal people once trod (which is everywhere people live) must be gone over with a fine-tooth comb and an artist's paintbrush to make sure no flint arrow-head is not overlooked. Then if any are found, the whole area must be regarded as sacred and untouchable...at least until a sufficiently large sum of cash is ponied up to square things with the spirit-gods.
Interesting article. I like how you separate out the planning and construction phases, as most people do not do that.
If you ignore delays due to government regulations, then I would expect the planning phase to go much faster in the modern world. A very small number of engineers with CAD software programs should be much more productive than traditional draftsmen, particularly in making iterative updates to plans. Plus we are constantly learning with each completed project.
I wonder if the apparent speed for both planning and building between ~1930 and 1950 isn't directly downstream of FDR's efforts to employ people during the Great Depression. I know the Army Corps of Engineers did a LOT of civic projects specifically to have the excuse to keep the boys employed and fed, but I'm not sure how much of that would have ended up in bridge-building rather than other, less engineering-intensive construction projects.
Less regulations and people willing to work hard and relatively cheap, plus deflationary pressures on material prices, must have helped the projects along faster. Complete opposite of the California High Speed Rail, the poster child for a system that seems to rely on Xeno's metaphor for goal setting.
I'm curious about the planning phase. I'm a PNWer, and it seems like large public projects here have several kinds of planning phases. One is engineering planning -- how are we going to build the bridge, environmental mitigation, etc. This is all largely civil engineering work. This doesn't seem to have become inordinately more complicated over time. Another is financing planning. What entities are going to pay for this, what is the split, what are the underlying bond and tax sources. I don't know if this has become more complicated, tho it seems like maybe it has. And then there is what I will call the "rights" planning stage -- whose property is this going to affect, and allowing all the legal challenges around that to play out. This seems like maybe it has become inordinately more complicated. Curious what you see.
This is an area where modern LLM should allow fast extraction of what used to be years (post-graduate theses’ worth!) of painstaking research digging into old newspaper articles and postmortem reports, as long as Claude and Gemini have ingested all the microfiche archives and paper books in American reference libraries (have they?)
I also suspect that “rights” (or lack thereof in the “bad old days” of ramming freeways through communities) are driving a lot of the increase in planning timelines. Should be provable with reference to sources as described above.
Complexity as the hidden variable in American infrastructure. Planning timelines haven't actually exploded the way the narrative suggests — the real constraint is organizational urgency, not regulatory capture alone. Potter's data complicates the simple story, which is why it's worth reading.
Separating planning from construction is the move that makes this comparison honest, and it also locates where the slowdown actually lives. Construction time is bounded by physics and capital, both of which have improved. Planning time runs on a procedural clock that no amount of money or technology speeds up, which is the same reason datacenter and grid buildouts keep missing schedule: the binding constraint is the interconnect study and the permit, not the steel or the silicon.
Bridges typically involve the institutional and political coordination of multiple jurisdictions since they, by definition, connect two areas. For example: two cities within a state, or two states, two counties, two countries, etc. Any sense of how this analysis would alter if you take into consideration the number of political jurisdictions directly connected to the bridge?
I strongly agree with your concluding sentence--it's crucial to have a profound sense of urgency about these kinds of projects. I lived in Los Angeles when the Northridge earthquake struck, collapsing overpasses, ramps, bridges and buildings. In the aftermath there was pervasive despair, and the assumption was that commutes would be FUBAR for many months and probably even years. Mayor Riordan was spectacularly successful in driving deals for the reconstruction and repairs, and the work started immediately and went on around the clock, three shifts, seven days a week. The construction companies were to pay a penalty if they missed their deadlines, and were to receive bonuses if the freeways opened ahead of schedule. I forget how much ahead of schedule things happened, but it was substantial.
The city then continued the momentum by reinforcing the undamaged bridges and overpasses throughout the city with steel cladding. I used to commute from Palos Verdes to Westwood and it was impressive to see this work happen without road closures.
It seems like a replacement might actually be harder to complete than a green field construction project. It's difficult to reroute traffic, do closures, etc. I'm fairly certain some amount of inefficiency is added by completing projects piecemeal to reduce the impact to drivers, etc.
I was curious how the Penobscot Narrows bridge could be planned and built so fast (planning and construction started 2003, finished 2006).
I'll quote Claude:
"In 2003, during a planned rehab of the Waldo-Hancock suspension bridge's main cables, MaineDOT found the cables severely corroded under their sheathing — bad enough that they cut the load rating from 100,000 to 24,000 lbs overnight and ran 16 temporary strengthening cables to keep it standing. Whereas a project of this magnitude normally takes at least a decade, this one went from concept to opening in 42 months. (Maine State Government)
The "how" breaks into a few decisions:
Design choice. They switched from suspension to a concrete cable-stayed bridge. Cable-stayed designs require less cable, can be built from identical precast segments, and are faster to erect. (Maine State Government) Precast segmental construction is the big speed lever — you cast deck segments off-site in parallel with foundation/pylon work, then erect them by balanced cantilever, so fabrication and erection overlap instead of being sequential.
Delivery method. The schedule killer in normal projects is the "design fully, then bid, then build" sequence. They used what MaineDOT called "owner facilitated design/build," partnering MaineDOT with FIGG as designer and Cianbro/Reed & Reed as contractor. (Wikipedia) Design-build lets construction start before design is 100% complete, and the owner-facilitated piece means MaineDOT was making decisions in real time rather than through formal change orders. FIGG was selected within weeks of the July 1, 2003 kickoff. (Maine State Government)"
The useful distinction here is that planning time is not just delay. It is where institutions decide which objections, studies, and uncertainties become part of the official project. Once that boundary changes, execution time becomes a misleading measure of capacity.
Did the bridges in the sample have comparable span and total lengths? One would assume new bridges display a strong survivorship bias towards increasingly challenging sites.
Great research, with this caveat.
I would have found the planning phase estimates to be more useful if they were devoted to the engineering/environmental related planning. I can't say for other readers, but when I think about how long a project is going to take I'm focused on the professional engineering planning (along with the construction period.) But using an announcement by politicians or bureaucrats at an agency as the start time isn't quite as clear cut as when the major professional contracts are let. When the civil engineers, and structural engineers are chosen and their clocks start, that could be a better start point. That would include enviro studies and surveys, but not include years of political fighting or zoning fighting or land purchase machinations. In sum, either when a competition is concluded and the designers are chosen or when a direct contract is established with the lead structural engineers, this approach would help limit what is otherwise a very open-ended idea of "planning" that leads to some of your listed projects going on for decades or a half century.
This approach would help avoid the future hypothetical example of the Strait of Messina bridge, which has been proposed for literally centuries, and which has had contemporaneous proposals since about 35 years ago. And even though some recent efforts have led to some engineering contracts, the reality is that the "planning" phase of the project hasn't started by my definition and only would do so when all the legal challenges are resolved and work is underway to do the detailed *design work* that leads to actual construction.
When people say "gee, it takes way to long to build anything", they're using a language of time, but I suspect really talking about money: "gee, we seem to be spending a whole lot of time and money before we actually put anything in the ground."
To that end, I'm not sure the planning metric is completely valid. "Let's take some extra time on the geology" is different from "let's spend millions saving some snail and moving 6 Swainson's hawk nests before we do anything." Technically they're both "planning" but there is a qualitative difference.
I'm not saying this to be critical of the latter behavior. Wealthier societies can worry more about snails and hawks; poor societies just need the stupid bridge built now. But I do think it makes your "planning" comparison somewhat less valid.
I didn't see the new Gordie Howe International Bridge between Detroit, Michigan USA and Windsor, Ontario CANADA. I believe it was scheduled to open for traffic sometime this month (perhaps it's already open). Do you have planning/construction times on this bridge?
Great article. I’m an environmental planner at a civil engineering firm. People sneer about birds’ nests and obscure plants but on most projects I’ve worked on the real resource sinks are historic/cultural resources (defined in an absurdly broad fashion IMO) and public engagement. The latter can trigger significant design changes, which of course cost a lot of time and money. Politics can also be tough for replacements that haven’t actually collapsed, especially when funds are limited. It’s a lot more fun for officials to announce new projects.
"Historical/cultural" in Canada means any land where aboriginal people once trod (which is everywhere people live) must be gone over with a fine-tooth comb and an artist's paintbrush to make sure no flint arrow-head is not overlooked. Then if any are found, the whole area must be regarded as sacred and untouchable...at least until a sufficiently large sum of cash is ponied up to square things with the spirit-gods.
Interesting article. I like how you separate out the planning and construction phases, as most people do not do that.
If you ignore delays due to government regulations, then I would expect the planning phase to go much faster in the modern world. A very small number of engineers with CAD software programs should be much more productive than traditional draftsmen, particularly in making iterative updates to plans. Plus we are constantly learning with each completed project.
I wonder if the apparent speed for both planning and building between ~1930 and 1950 isn't directly downstream of FDR's efforts to employ people during the Great Depression. I know the Army Corps of Engineers did a LOT of civic projects specifically to have the excuse to keep the boys employed and fed, but I'm not sure how much of that would have ended up in bridge-building rather than other, less engineering-intensive construction projects.
Less regulations and people willing to work hard and relatively cheap, plus deflationary pressures on material prices, must have helped the projects along faster. Complete opposite of the California High Speed Rail, the poster child for a system that seems to rely on Xeno's metaphor for goal setting.
I'm curious about the planning phase. I'm a PNWer, and it seems like large public projects here have several kinds of planning phases. One is engineering planning -- how are we going to build the bridge, environmental mitigation, etc. This is all largely civil engineering work. This doesn't seem to have become inordinately more complicated over time. Another is financing planning. What entities are going to pay for this, what is the split, what are the underlying bond and tax sources. I don't know if this has become more complicated, tho it seems like maybe it has. And then there is what I will call the "rights" planning stage -- whose property is this going to affect, and allowing all the legal challenges around that to play out. This seems like maybe it has become inordinately more complicated. Curious what you see.
It's hard to tell because so few bridges have easily available information at this level of granularity.
This is an area where modern LLM should allow fast extraction of what used to be years (post-graduate theses’ worth!) of painstaking research digging into old newspaper articles and postmortem reports, as long as Claude and Gemini have ingested all the microfiche archives and paper books in American reference libraries (have they?)
I also suspect that “rights” (or lack thereof in the “bad old days” of ramming freeways through communities) are driving a lot of the increase in planning timelines. Should be provable with reference to sources as described above.
Complexity as the hidden variable in American infrastructure. Planning timelines haven't actually exploded the way the narrative suggests — the real constraint is organizational urgency, not regulatory capture alone. Potter's data complicates the simple story, which is why it's worth reading.
Separating planning from construction is the move that makes this comparison honest, and it also locates where the slowdown actually lives. Construction time is bounded by physics and capital, both of which have improved. Planning time runs on a procedural clock that no amount of money or technology speeds up, which is the same reason datacenter and grid buildouts keep missing schedule: the binding constraint is the interconnect study and the permit, not the steel or the silicon.
Bridges typically involve the institutional and political coordination of multiple jurisdictions since they, by definition, connect two areas. For example: two cities within a state, or two states, two counties, two countries, etc. Any sense of how this analysis would alter if you take into consideration the number of political jurisdictions directly connected to the bridge?
I strongly agree with your concluding sentence--it's crucial to have a profound sense of urgency about these kinds of projects. I lived in Los Angeles when the Northridge earthquake struck, collapsing overpasses, ramps, bridges and buildings. In the aftermath there was pervasive despair, and the assumption was that commutes would be FUBAR for many months and probably even years. Mayor Riordan was spectacularly successful in driving deals for the reconstruction and repairs, and the work started immediately and went on around the clock, three shifts, seven days a week. The construction companies were to pay a penalty if they missed their deadlines, and were to receive bonuses if the freeways opened ahead of schedule. I forget how much ahead of schedule things happened, but it was substantial.
The city then continued the momentum by reinforcing the undamaged bridges and overpasses throughout the city with steel cladding. I used to commute from Palos Verdes to Westwood and it was impressive to see this work happen without road closures.
It seems like a replacement might actually be harder to complete than a green field construction project. It's difficult to reroute traffic, do closures, etc. I'm fairly certain some amount of inefficiency is added by completing projects piecemeal to reduce the impact to drivers, etc.
I was curious how the Penobscot Narrows bridge could be planned and built so fast (planning and construction started 2003, finished 2006).
I'll quote Claude:
"In 2003, during a planned rehab of the Waldo-Hancock suspension bridge's main cables, MaineDOT found the cables severely corroded under their sheathing — bad enough that they cut the load rating from 100,000 to 24,000 lbs overnight and ran 16 temporary strengthening cables to keep it standing. Whereas a project of this magnitude normally takes at least a decade, this one went from concept to opening in 42 months. (Maine State Government)
The "how" breaks into a few decisions:
Design choice. They switched from suspension to a concrete cable-stayed bridge. Cable-stayed designs require less cable, can be built from identical precast segments, and are faster to erect. (Maine State Government) Precast segmental construction is the big speed lever — you cast deck segments off-site in parallel with foundation/pylon work, then erect them by balanced cantilever, so fabrication and erection overlap instead of being sequential.
Delivery method. The schedule killer in normal projects is the "design fully, then bid, then build" sequence. They used what MaineDOT called "owner facilitated design/build," partnering MaineDOT with FIGG as designer and Cianbro/Reed & Reed as contractor. (Wikipedia) Design-build lets construction start before design is 100% complete, and the owner-facilitated piece means MaineDOT was making decisions in real time rather than through formal change orders. FIGG was selected within weeks of the July 1, 2003 kickoff. (Maine State Government)"
Here's a picture of the old and new bridges side by side: https://media.newscentermaine.com/assets/NCM/images/367b2c96-5805-404b-a7b2-98d37a77d34b/367b2c96-5805-404b-a7b2-98d37a77d34b_1920x1080.jpg
How to Replace an Interstate Bridge in 48 Hours
https://www.youtube.com/watch?v=tq_pEP0yK2Y
The useful distinction here is that planning time is not just delay. It is where institutions decide which objections, studies, and uncertainties become part of the official project. Once that boundary changes, execution time becomes a misleading measure of capacity.