Government
Design-build may look like lack of planning, but isn't
Is DDOT under Gabe Klein dismissive of planning, throwing caution to the wind as it builds a bike lane here and a streetcar line there? Or does this objection reflect an outdated view of how project planning should be done?
Image via Benchmark Construction.
Gabe Klein comes from the business world and has said several times he wants to run DDOT like a business. To understand how he might do that, we should understand what businesses have learned about good project planning.
The traditional, 20th-century view is that project managers hold endless meetings with stakeholders to gather requirements and create a detailed plan as the first stage of any project. This "Big Design Up Front" method went by various names Project managers in various industries have adopted a leaner approach in which the build phase begins before a comprehensive plan is completed. In software, this is becoming known as the "minimum viable product." A single team of designers and developers designs and builds a product, and seeks to get that in front of users as soon as possible. That works because no amount of feedback to a paper plan is nearly as valuable as observing how people actually use a product. In construction and transportation, this leaner approach is known as Design-Build.
Design-build creates a single team of designers and builders whose constant communication enables certain build stages to begin before the plan is complete. In design-bid-build, designers create a complete plan and then "throw it over the wall" to builders who hadn't been involved in the design, creating two groups who point fingers at each other when problems arise. In contrast, a design-build team forms a single point of accountability.
Furthermore, the expedited construction means that total costs are lower and feedback to a preliminary working system is received far earlier, while there's still time to modify the plan. This in turn leads to higher quality. In fact, Design-Build was used in the rebuilding of the I35W bridge in Minneapolis.
DDOT under Klein has increased the adoption of design-build, implementing the "Every Day Counts" initiative from the DOT to reduce the duration of transit projects from the current national average of 13 years, which Klein says is "unacceptable". The $300 million 11th Street Bridge replacement in Anacostia is the largest DDOT project to date to use Design-Build.
That's not to say DDOT shouldn't listen to people. Startup guru Eric Ries, who popularized Minimum Viable Product for tech startups, also talks about the importance of getting feedback from customers. The big difference is that the feedback doesn't all happen before work begins on the product.
In a tech company, programmers can build a fairly rudimentary version of the product and release it for users to try. DDOT can't release a rough version of the bridge for drivers to try, but what they can do is to share the designs as the project evolves. With the 11th Street Bridge, DDOT and the contractor may be making changes as they go, but they haven't been as proactive about getting feedback from potential drivers and residents along the way.
The same is true for the streetcar. The use of the initial streetcar line is essential input into the design of the rest of the streetcar network, far more meaningful than feedback to the 300-page plan that DDOT circulates for stakeholder feedback. As items come up, like the placement of the maintenance yard or technology for minimizing overhead wires, DDOT is adapting to those. Furthermore, the use of the initial streetcars and responses of drivers is also invaluable to DDOT's traffic engineers. No amount of computer modeling of ridership and driver reactions to streetcars is as informative as observing the actual use and reactions to a streetcar line, because traffic engineers aren't modeling water in a pipe, they're modeling people.
Still, until the streetcar is built the "product" is a set of plans which riders need to know about to give input on and help the project evolve. That's why DDOT needs to maintain constant communication about these issues and share the evolution of the project through the Design-Build process.
So far, with the streetcar, they are. DDOT is releasing a lot of details and plans about the streetcar. The main objection coming back is that people expected all of these details to be decided before a single dollar was spent, and in fact for everything to be decided in advance about the entire 20-year vision. By not doing so, Gabe Klein isn't being irresponsible. Instead, he's creating a more agile DDOT that may be a model for other transportation agencies going forward.
The outdated approach to planning.
Ken Archer is CTO of a software firm in Tysons Corner. He commutes to Tysons by bus from his home in Georgetown, where he lives with his wife and son. Ken completed a Masters degree in Philosophy from The Catholic University of America. 
Comments
@ggwash
- Cyclists are special and do have their own rules
- M Street cycle track keeps improving, draws church anger
- O'Malley announces first projects using new gas tax money
- Can Loudoun grow while protecting its rural areas?
- ICC losing bus service in classic bait and switch
- Silver Spring mall could get massive facelift, new name
- WMATA launches "Short Trip" rail pass on SmarTrip
by Brian on Jul 14, 2010 12:33 pm • link • report
The version of the tire swing comic I'm used to is a bit different, but has the same idea...
On the other hand, the "Big Design Up Front" method can have its advantages for huge infrastructure projects. Metrorail got built almost exactly to specifications, and has been fairly successful. Had the "Design-Build" methodology been used, I'm pretty certain that the Green Line would have never been built. In many ways, 21st-century DC's been built around Metrorail, rather than vice-versa.
Total failures of the "Big Design" methodology can be horrific too. Although I'm very glad that DC does not have 3 beltways as was initially proposed, the bits of the original plan that did get built make no sense. Because there was no mechanism in place to pragmatically scale back the freeway network, many projects were halted, and we were left with a bizarre stub of 395 running through downtown, NY Ave became a de-facto freeway, and more than half of the Southeast Freeway serves no meaningful function. The 11th St Bridge project should hopefully correct many of the SE Freeway issues, although it took nearly 50 years for it to do so.
by andrew on Jul 14, 2010 12:40 pm • link • report
Am I the only one who *likes* the new traffic pattern there? The signal timing could use some work, and they need more signage. However, I feel *MUCH* safer driving through that intersection and making turns.
Now that the new landscaping is going in, it's remarkably starting to look rather nice as well.
by andrew on Jul 14, 2010 12:41 pm • link • report
you may feel safer driving - try walking through it.
by Brian on Jul 14, 2010 12:50 pm • link • report
It sucks. The graphic up top is laughable. Establishing cost at the preliminary design stage is a recipe for disaster for the client. I assume for public projects that is required to be a "bid stage", where several companies try to establish cost based on incomplete design. That just increases the chances for cost mayhem during the latter stages due to wrong assumption, low-balling, etc.
Just from the designer's standpoint, we feel like to fully serve the client, from good design all the way to regulatory advice, we need that unfettered relationship to exist during design, before the cost/schedule driven contractor comes into the picture.
Also, during construction it is the role of the designer to QC the contractor's work in the field, to protect the client and make sure the work conforms. When you put the designer on the same side of the table as the contractor, there's a conflict created.
I know those points don't all apply to all industries, but you really need to be through the process to see the whole picture.
by Lou on Jul 14, 2010 12:56 pm • link • report
The rest of everything that goes on is a design-build task force of the built environment. This task force is made up of developers, lawyers, bureaucrats, financiers, citizens, and architects. The "planners" are really just responsible for statutory land development review as part of the ongoing, never-ending, iterative design-build project to create an ever-changing built environment.
Shouldn't we change the title of city planning to Visioning and Statutory Land Development Review? Yeah, I guess it doesn't quite sing, does it?
by Dave M on Jul 14, 2010 1:23 pm • link • report
by IMGoph on Jul 14, 2010 1:34 pm • link • report
I gotta agree with @andrew. Don't see any problems with that re-routing. Also, a snort is not an argument. :)
by oboe on Jul 14, 2010 1:57 pm • link • report
Pretty much the only thing we know at this point is that, some day, there will be a streetcar on H Street. Proceding from that point, DDOT decided to install the rails while the entire streetscape was being torn apart completely and put back together.
In doing so, they saved (pardon my French) a shit-ton of money, and in ensuring the road would only be torn up once, did H Street merchants a huge favor.
But, by all means, let's wait to drive even a single nail until every single specification deliverable for the entire multi-billion dollar city-wide system has been formalized.
by oboe on Jul 14, 2010 2:06 pm • link • report
In brief, if Klein is really trying to apply commercial rules to government procurement, then he really doesn't understand how government has to work ... if it is to be responsible government.
by Lance on Jul 14, 2010 2:31 pm • link • report
Interesting ... cause I understand that some government mandated environment impact studies were overlooked in the process AND, as many bloggers have mentioned here, there's no certainty that there's room for streetcar rails on that bridge ...
by Lance on Jul 14, 2010 2:35 pm • link • report
And there's plenty of room on the local bridge for streetcar rails...
by Froggie on Jul 14, 2010 2:46 pm • link • report
Knowing I am a pariah among the crowd, feel free to ignore the following, but design-build is the foundation of my consultancy and I have directly managed more than 3 billion of construction design-build the past 25 years, my company, many times that.
Design-build is leaner only because of lack of competitive bidding post design, and the “ever” so slight reduction (1-3%) in design time (mostly as a function of qualifying numerous potential designers) as result of and the abbreviated time frame that results. You are sorely mistaken if you think the design-build process streamlines the feedback and “study” phase one bit. Endless design charrette's, followed by months, if not years of stakeholder input and plan revision are typical of design build, just as it is in design-bid-build etc.
“Design” is the primary foundation of design-build, and the criticism that DDOT is rightfully getting is that there has been no “design”, thus far, only pipe dream power points presentations. No way to power the system, no way to connect the H Street part to Union Station or Anacostia, no funding, no guesstimates at ridership and consequently, no budget numbers anyone would realistically put any stock in. They keep saying 1.5 billion, but I would promise ownership of my company that if this ever happens, it will be a minimum of twice that number.
You support this method of “build it and see” by saying…”No amount of computer modeling of ridership and driver reactions to streetcars is as informative as observing the actual use and reactions to a streetcar line, because traffic engineers aren't modeling water in a pipe, they're modeling people.”
100% wrong. DDOT hasnÂ’t even given back of the napkin numbers yet for proposed ridership, but we will forget about that for a second.
People in DC aren’t any more unique than people in SF, Portland, Miami or Boston and this really isn’t rocket science. Streetcars aren’t “theoretical”. Ridership numbers are based on a variety of “set” criteria that transportation analysts and engineers have refined over the years on similar systems around the country and world. DDOT doesn’t have the in-house talent to take this task, but that’s no reason to ensure it isn’t done properly. They could hire the outside expertise they need to arrive at a number, then decide what kind of margin of error is acceptable to them…say ~10%, and voila, an actual, honest to god factually based “number”.
Design-build has some pretty significant draw backs, cost being the primary one, especially as it related to publicly funded projects. Once you’ve entered into the AIA contract with a design-builder, you’ve locked yourself into their pricing structure, when you would instead typically find gross savings of anywhere from 5-15% had you competitively bid the project. You sacrifice higher cost for the “hope” of a reduced schedule. It doesn’t always work that way.
Lastly, my critique of Gabe Klein. While I am sure he is a nice guy and while I am sure he did a great job at Zip, he is in no way qualified to run a 180 million public dept of transportation. Regional VP for a car rental company? Budget Rent a car has/had more rental cars in DC than did Zip when he was there, yet you would all cringe at the thought that the DC Manager for Budget Car Rental was in charge of a multibillion dollar heavy-highway/public transit project.
After that he ran a dc catering business to compete with the hotdog stands on DC streets. Catering company = head of DDOT? Really? I am not saying local government couldn’t benefit from a bit of private business experience, but it is no reason to ignore all sane conceptions of “qualification” and “experience”.
by nookie on Jul 14, 2010 3:06 pm • link • report
Read the data on the ddot website http://ddot.dc.gov/DC/DDOT/About+DDOT/Publications/DC+Transit+Future
This will be the busiest transit line in the city according to the report. How much modeling do you want done? Another few million $ worth?
Whether Gabe Klein ran 1 company, 5 companies (I think it's about 4-5). What does that have to do with anything? Zipcar is going public last I heard.
Design-build is what most states are moving to for large scale infrastructure projects, and PPP's. I for one am glad the new DC is not terribly behind the times as it was for so long. DC is leading the way now. Nookie, Lance etc. are what we call "haters" in the hood. Never have anything good to say. Mind made up before they even read the post.
by dontedc on Jul 14, 2010 3:21 pm • link • report
hmmm .. it sounds like you didn't read the prospectus? Zipcar is going public because it's profits are so bad that the current owners of its stock are doing their damnest to get their money out of it before it goes belly up. I.e., anyone actually buying that stock will be left 'holding the bag'. We, the taxpayers of DC, can't afford that kind of business model for our transportation systems.
@nookine, please tell me you're joking about the hot dog stand 'experience' ... ?
by Lance on Jul 14, 2010 3:27 pm • link • report
by Lou on Jul 14, 2010 3:40 pm • link • report
Wow, a "plan" that "just" came out two months ago. As folks have been saying, this streetcar thing has been in the works for 11 or 12 years. At a minimum this "plan" is about a decade late.
They say they expect the feds to pay for 25% (oops, we just saw last week that assumption is already wrong and we are already behind on that guesstimate) and that 50% will come from "Corridor Specific" contributions. What? What are those contributions, how did you arrive at that number, what basis do you have for getting to that number?
They come up with a number for yearly cost to operate and maintain (68 mil). They've oddly done that without identifying a proposed fare structure, so that number is a complete "out-of-ass" number.
Oh, but they make it all better by immediately saying...
"Project operating and maintenance costs not covered by
fare box revenues are assumed to be covered by a local
funding contribution."
I.e...the DC General Fund.
Well no joke, if you don't get your imagine money from the non-plan you have for the Feds, the non-plan you have for the corridor specific funding, and the non-plan you have for fares, then OF COURSE the general DC taxpayer will have to fund it.
I know this colorful pdf has wowed you, but it is as devoid of actual, needed, fundamental data as the one they put out last year,. I know you can't see that, I'll chalk that up to lack of experience in such things, but the rest of us are petrified that a city currently half a billion in the hole, that already had a broken, relatively new mass transit system it can't afford, will be on the hook forever for another multibillion dollar, unfunded system.
@ Lance, I wish I was kidding...quoted from the Washington City Paper 12/31/08 edition...
"Since he left Zipcar, he's headed up On the Fly, the vending and mobile catering business that's helped end the street-vending monopoly held by the half-smoke crowd. (He also did the catering for Fenty's birthday party/fundraiser earlier this month.)".
Pure, unadulterated ridiculousness.
by nookie on Jul 14, 2010 3:51 pm • link • report
Another example of design-build in DC is Nationals Park. Personally, I think the design missed a lot of great opportunities. But they got it done fast. This is where there's a lot more subjectivity on the role of design in making a great place, as opposed to building the plans quickly and competently.
There's also the scope of what's being done via design-build. Andrew mentioned how if Metro had been done design-build, the Green line never would have happened. That all depends on the structure of the contracts and the scope of the design work. For streetcar, all the planning is being done for the system as a whole, while the nitty gritty engineering decisions for each individual segment won't be decided upon until it's time to do that segment. That same basic approach applied to Metro, too - they didn't know where the individual power substations for the last Green Line segment (completed in 2001) were going to go in 1973 when the oldest parts of the Red Line were under construction - yet this is exactly what some people have demanded as a part of the 'planning' for streetcars.
by Alex B. on Jul 14, 2010 3:56 pm • link • report
Now this really can't be true, can it? What kind of Banana Republic are we living in?!?!
by Lance on Jul 14, 2010 4:55 pm • link • report
No ... we want to know where the power substation is going to be for the part of this system goes up ... And how it's going to go through a tunnel that the owners of said tunnel have said 'no'.
by Lance on Jul 14, 2010 4:59 pm • link • report
by Paul Carlson on Jul 14, 2010 6:28 pm • link • report
by Transitguy on Jul 14, 2010 6:37 pm • link • report
Similarly, design-build can (and hopefully will) be used to correct some of the pedestrian-related deficiencies of the Dave Thomas Circle. For one, the crosswalk across O St definitely needs better signage, and possibly even a pedestrian-triggered traffic signal. Lighting could also use some major improvement here.
by andrew on Jul 14, 2010 9:30 pm • link • report
I realize money comes out of general fund to cover overages in our obligations. What YOU don't realize is that is the absolute 100% wrong way to pay for something. Needing to steal money out of the rainy day fund to cover your basic yearly budgeted obligations only proves how little forethought or planning went into it in the first place. And now you are actually proposing...no SUPPORTING that rather than DDOT do any basic level of financial due diligence on how to pay for ~700 million dollars of O&M costs over the first decade alone, they should just plan on taking it all out of the general fund!! Epic fail my friend...it's obvious you've never even bothered to balance your own checkbook.
Live in whatever fantasy world you prefer thinking that everything is rosy and fine. And yes, I am quite confident that have more "transit" experience (from concept, funding, design and construction) than probably everyone who has posted on this topic combined. Heck, the track and tunnel design and construction management I provided for the dulles aerotrain the past 7 years alone probably makes me more qualified than anyone at DDOT on "transit" but again...whatever fiction you prefer that allows you to sleep better at night.
by nookie on Jul 14, 2010 9:56 pm • link • report
The
by RegBazile on Jul 14, 2010 11:50 pm • link • report
@Lance, please go back and read @AlexB's post at 3:56, as it addresses this directly. There are large-scale engineering (software and otherwise) projects for which what you say is true. There are others for which it most certainly is not.
by oboe on Jul 15, 2010 10:07 am • link • report
Wow, I am going to stop reading this blog as much as I hate to say it. The negativity and venom from unhappy people in the comments ruins it for me. I just want factual information. Not a bunch of obviously over the top, disparaging remarks.
Welcome to The Internet! You must be new here... ;)
by oboe on Jul 15, 2010 10:08 am • link • report
Add a Comment