Virtual Stringline: Common Themes on 3D GPS Modeling
by Harry O. Ward PE
September 1, 2007
I just completed a two-part series on one of the more interesting issues that I have come across in my career, and received quite a bit of mail about it. Many states have begun to mandate that a line of work that previously did not require a professional license--namely, 3D GPS modeling--now requires a stamp of approval from a professional engineer or licensed surveyor.
This requirement involves engineers, surveyors and contractors, and I think Site Prep identified this so early in the game that many people in our respective professions were not aware of it spreading across the nation. I have spoken with professionals who admitted that they were embarrassed to learn that their state passed this directive without them knowing about it, and these were people who strive to stay on top of things.
I received many distinct perspectives on the topic (read some reader responses here). Some respondents--mostly licensed individuals--were highly in favor of these position statements. Others--mostly unlicensed people in the data prep business who believe their livelihoods are being threatened--hold the opposing view. Each of these perspectives is completely understandable. Let me share the common themes I noticed from the many responses I received.
Reviewing Parts One and Two
In Part One of my column in the May/June issue, I discussed new laws on licensing and interviewed the executive directors of some of the affected states. But these “new laws” aren’t really new laws at all; they are actually position statements that declare that data prep now falls under the definition of civil engineering and surveying work and, as a result, a professional licensee is now required to certify that work. What it comes down to, we learned, is that the boards are really requesting someone to be accountable for this work--legally, professionally and financially.
In Part Two of my column in the July/August issue, I discussed the situation from different contractors’ viewpoints. Some contractors believe these regulations don’t really affect them in a negative way because their firms have already been hiring PEs or LSs to ensure proper quality of their data models. In fact, they now believe that they have a business right and a professional mechanism to pass this cost on to the developer/owner. The other camp believe that since they are licensed contractors in the state, by that very nature they should automatically inherit any liabilities involved in the construction. As a result, they don’t believe that another authority should have to oversee this work.
Common Themes
One extra measure of public protection that licensing affords is that not only are legal matters held accountable to someone, but ethical matters are addressed as well. Businesses that do not have professionally licensed engineers/surveyors are not held to the same level of ethics violations as PEs and LSs are. Engineers and surveyors can lose their right to practice if found guilty of ethical violations.
Many comments from both licensed and non-licensed senders involved the argument that being licensed does not mean an individual is actually competent in this area of expertise. The common theme is that the vast majority of engineers and surveyors wouldn’t know a good digital data model if they stumbled over it. Engineers took the brunt of these comments because surveyors have been preparing stakeout data forever from the engineer’s “pretty pictures” and are well accustomed to the ambiguities that engineers allow in the designs. As both an engineer and a contractor, I am split professionally. But, I do think engineers should place more thought into the constructability of their projects than they currently do.
As an example, I was recently asked to build a 3D model from a contour file provided by a design firm for a 300-home subdivision and golf course. The project was to be built using 3D GPS robotics. In the file, there were cul-de-sacs where only one contour crossed the cul-de-sac, with no surrounding spot shots to help define the surface intentions. Yet the contractor was supposed to build the road to a 3/8-inch tolerance. Also, there were contours sequentially drawn at 2-foot intervals that skipped various elevations. For instance, the contours would begin in an area at elevation 430, the next one was at 432, the next was at 438, the next was at 440 and so on. Naturally, one would inquire as to what happened to the contours at 434 and 436 and how that would then impact the rest of the site.
My team requested more information for this and other sparse data areas. We were told that no more budget existed on that project and that the data we had was all we would receive. We were told to do whatever we thought best to move forward. We did just that and the project is now built and open to the public. Although that design firm’s attitude was shameful, at the same time it was perfectly honest.
From readers I received a lot of examples of poor work in CAD files sent to contractors that impede building. Some include:
- Contours that have elevation labels but no physical 3D elevations.
- Layers misnamed or worse yet, numbered layers without any name.
- Multiple layers all depicting proposed contours in various stages of completion yet without guidance as to which should be used.
- Contours and building pads floating in 3D at elevations different than the elevation labels.
- Roadway profile data that conflicts with the cross sectional drawings.
- Annotation callouts for grading slopes and percentages that conflict with the contours and spot shots identified.
At Cross-purposes
The issue that arises here is that we are at cross-purposes. The engineering or design firm obtains a contract to prepare a set of plans that will be reviewed and approved by the local jurisdiction(s) and agencies. Those plans are then bid out to contractors for construction. Once awarded, unfortunately for the owner/developer, those plans are often not completely constructible and fees must be paid to fix shortcomings.
Engineering and design firms use contour data and selected spot shots to depict what a surface should be defined by. This is suitable for reviewers to see the intentions of the surface grading, but it is usually unsuitable for construction. Contractors, and specifically 3D GPS models, require breaklines and points.
So I am going to go out on a limb here and declare in writing what I have been teaching for years: Contours are output, not input. Engineering and design firms must stop redlining contours onto sets of plans and digitizing them into the CAD system. This merely gives an idea of what the grading of the site will accomplish. It does not provide the accuracy to achieve a 3/8-inch as-built tolerance check.
When contours are used as input to the 3D model, then straight-line interpolations guide the actual model. Except for building foundations, few things are totally planar on our sites. The entities that should guide these models are breaklines, which are developed according to the elevations, slopes and grades desired by the designer. These objects are the sources for the digital terrain model, which then produces contours. These breaklines and points will develop the precise surface that 3D GPS works best with. I call these “millimeter accurate” models. If the model is accurate to the millimeter, the machine control equipment has a much better chance of constructing the site to a 3/8-inch tolerance. Hence the contours are output from the digital model; they are not the input that creates that model.
The sooner engineering and design firms learn this and embed it into their staff’s skill sets, the entire family of industry comprising civil, survey and construction will benefit. When laying out digital models for proposed sites, the source data has to be the very essence of the intent--that is, elevations, slopes and grades.
It Is Not All About the 3D Model
What I have been trying to convey over several years, several articles and many speeches is the following: It is not enough that a correct 3D model is prepared. Although this is a good start, the engineering or design firm must be compensated for the additional effort it takes to prepare a 3D model to be used for construction versus a 3D model to be used for review and approval. Further compensation is required for accepting the liability for the model. When the owner requests that a correct 3D model be provided to the contractor, there should be a fee associated with that transfer. It should incorporate payment for the precision and correctness of the model along with a payment to accept liability for anything incorrect within the model. I wouldn’t be surprised if that fee doesn’t approach a full-blown stakeout fee. Why? Because surveyors have performed similar work for years, and they take responsibility for the quality and accuracy of the original stake placement. They have also been paid for doing so.
Additionally, there should be a clause in the transfer agreement that takes into account the ownership of this intellectual property and what rights or lack thereof pass to the recipient. Consider the following:
- When I produce a digital stakeout suitable for use in 3D GPS machine control equipment, am I giving the rights to simply use my work product? That is, I may only be allowing that data to be placed into the robot’s controller and used for grading directly.
- Am I giving the right to modify my property? When change orders or other site-based observations result in a need to modify my data, have I provided that right? If so, there may be a fee involved. In another circumstance, if an engineering firm designs a site that gets approved and provides those drawings to the contractor, does the contractor have a right to modify those plans? Not usually since they are often copyrighted.
- Am I giving limited rights to modify my property when change orders ensue? And if so, do I dictate how many revisions may occur before it reverts back to me with the idea that too much has changed and now could cause a public hazard?
Standards for the Future
As we wind up this discussion, it appears that the public will hold those who build 3D GPS models to a higher standard than in the past. This is now a growing part of the field of civil engineering and surveying. The natural provider of this data is the engineering or design firm itself, but it is clear that the design staff need to awaken to 3D design, layout and a close synchronicity between the two. In the meantime, data prep firms or the contractors themselves will need to build these models for use in the growing technology of 3D GPS machine control equipment.
State boards have reported an increase in the number of serious modeling problems on projects in the past few years. Several engineering and design firms that have professional licensed individuals do not have staff who can create an accurate 3D model. Therein lies the issue. So the developer can do one of two things, or maybe both: 1) Insist that the engineering or design firm have the skills needed to create a correct 3D model that directly is linked to the creation of contours for the site, and 2) Use a data prep company that has licensed people overseeing and taking charge of the 3D model.
Perhaps when the boards are taking these positions they should consider adding a 3D grading and 3D GPS adjunct to their licensing exams. Another idea is to add another designation to the surveying or engineering licenses, something on the order of a Class C license that would be applicable to 3D grading.
Harry O. Ward PE
Harry
O. Ward, PE, is president of Harken-Reidar Inc. He has been a member of the
engineering faculty at George
Mason University
since 1997 and was named “Outstanding Adjunct Professor” for GMU in 2010. He
can be reached at hward@harken-reidar.com or hward@gmu.edu.
Comments (0) Post a Comment
Did you enjoy this article? Click here to subscribe to the magazine.