Data prep is rapidly maturing, and I discussed several aspects of the business recently with Bruce Flora, LS, a former competitor of mine and owner of Flora Surveying Associates PC and Data Pro Ltd. in Saluda, Virginia. Flora and I have both practiced the data prep technology/art for many years, and we discussed how data prep firms can provide additional services, seek out potential revenues and offer accompanying savings for clients.
|
To be successful, a data prep company must be set up wisely. In my experience performing data prep, I learned that although the tools and technologies were similar to the engineering I had done before, the actual business and related services were unique to construction and came with different challenges. It isn’t enough simply to know how to build a 3D model—you must structure your business well around your modeling services.
For example, Flora created a second company in 1998 to perform data prep. Although he already had his surveying company registered as a professional corporation, Flora realized he did not need a surveying license to perform data prep in the state of Virginia. As a result, he created Data Pro Ltd. as a separate entity so the liabilities inherent with data prep would be protected as a separate firm with limited liability. Nowadays Flora is comfortable with the business and technologies involved with data prep, so he no longer sees the need for this distinction and is considering merging it all back into the survey company. But the point is that data prep companies must be properly structured to deal with the demands of this business.
Staffing for Data Prep
A data prep company should be staffed with people who have strong experience in construction. The industry is just starting to understand the importance of construction surveying in making 3D models, and the best person to create these models is someone who has been pounding hubs and stakes for years. Flora relies on staff members who have strong surveying backgrounds, and he also considers construction engineers to be good candidates for this work. “The design engineers who are good at developing plans for review and approval may have a hard time understanding the construction aspects because they don’t get enough field experience or mentoring in their jobs as designers,” he says, adding, “and there are very few schools teaching it at this point.”
Because CADD was initially developed as a drafting tool, designers became ingrained in the use of CADD for drafting and have had a difficult time transitioning to 3D design. The career path of design engineers normally moves them up to management so they spend little time becoming expert in CADD, the tool of the trade. On the other hand, young design engineers may be familiar with computer technologies but not have the experience to design yet. Flora mentions that some construction firms have a company policy that an engineer must spend a year in the field before working in the office. The idea behind this policy is that the engineer must learn to build the project before he or she is fully qualified to design it.
This leads to the question of whether data prep has outgrown the engineering field. Or, in other words, has data prep matured to the point that it requires a special skill set that is beyond the traditional engineer? According to Flora, owners and developers now focus on regulatory approval, which does not guarantee good design. “Is it really practical that the engineer is responsible not only to get the job reviewed and approved, but to have it be constructible as well?” he asks. Flora sees nothing wrong with dividing and specializing roles so that the engineer continues with design and approvals while the data prep expert focuses on constructability.
Companies that try to combine these two roles often face business-related pressures that can cause them to lose focus. Flora tells of a meeting he had with a developer who was intent on making a deadline to present before the county commissioners and win project approval. The developer put pressure on the designer to get the job done in time to meet with the council, but of course something had to give—and that something was the quality of the work. The meeting deadline was met, and the job was approved. But when the project went to construction, it was quickly identified that the design wasn’t constructible. So the data prep arm of the project went back to the developer and pointed out that more money would be needed to complete a constructible design. Of course, the developer had a short memory of the pressure he had placed on the designer to make the deadline, which led Flora to call him the “Great Forgetter.”
It is important to understand that staff members’ training and differing levels of experience can affect the quality of the data prep. I have often said that I can train an experienced engineer to do CADD in about three days, but I need at least two weeks to train a CADD person to do engineering. I am only kidding, but this cuts to the point: If you don’t know the field, you have little business doing CADD work in it.
Establishing a Standard
When machine control was first introduced to the market, the pioneering experts built the models. These individuals lived and breathed GPS, machine control and 3D modeling. They saw the whole picture and helped establish the industry. But the industry is changing now that more data prep shops are opening, and some poor models are going into construction.
“Machine control manufacturers are becoming concerned because it throws the credibility of their equipment into question when something goes wrong in the construction,” Flora says. “Some now demand that an expert build the first model for new customers who purchase their equipment to ensure that they get started on the right foot.”
Flora was once somewhat against licensure for data prep, but he now sees it as a positive move to ensure that experts oversee the creation of these valuable models. As an example, Flora cited a recent highway project where the developer chose to save $3,000 on the creation of a model for the road construction, but the model was poor and caused $30,000 in rework that could have been avoided.
“This technology is now a two-way street with all of the magic and mystery gone. Machine control is now a tool of the trade instead of an oddity. Knowledge is now the critical aspect of the job,” Flora says, adding, “and the knowledge of the individual is now paramount.” I agree with Flora on this point because one of my mantras is “It is all about the data.” And, of course, a skilled, qualified individual must create that data.
Offering Advanced Services
Anyone who chooses to enter the data prep field must understand the expectations for the design plans, the resulting 3D model and the equipment used to build the job. Each component has goals, projected accuracies and time frames for completion, and each component also has limitations that should be established ahead of time.
The highly experienced data prep provider is able to look at the entire project from a macro level. This perspective encompasses the plans provided, the interpretations that must be made to complete the model, the equipment being used on the project and the local standards that are applicable. For example, Flora was involved in a parking lot project that was switched from asphalt to concrete due to the high cost of the asphalt. The site had a very complex grading plan; the parking area had smooth contours grading to internal drainage systems instead of cross site sheet flow. The machine to be used on this site had a 14-foot straight-edged screed, but Flora identified that this wasn’t suitable for a curved surface. As a result, he made some adjustments to the model so that it could be used to maintain design intent without affecting drainage areas.
The high-level concept behind this example is that the machinery, the standards and the model must be synchronized for the project to be a success. The field of 3D modeling and data prep now allows for a more sophisticated analysis prior to building. By focusing on constructability, data prep providers can build the model to serve the express purpose of the machines available to do the grading. I see this synchronicity as a sign of the maturity of the field of data prep.
In another case, a well-thought-out model was developed and used for a dense subdivision. The model included rough grading of house pads and surrounding landscape berms. Following this rough grading, the finished grading for house pads had to occur. Due to the tightness of the job, the contractor couldn’t stockpile the earth in the road or on adjacent lots since the lots were being sold and needed to remain pristine for the customers. When the contractor began building the home, he stockpiled the earth on the berms that were accurately sculpted into place, thereby destroying them. Summing up the lesson learned from this project, Flora says, “Aesthetics is important during construction, and import/export of material is critical. The site needs to be modeled in a way that reduces grading twice and leaves enough dirt in the pad so that final grading can occur while minimizing import and export. We find that the cost of construction—the fuel and labor costs, wear and tear, operations and maintenance—is driving this efficiency, and anything to minimize these costs is valuable.”
Another of Flora’s projects in New England required extensive boring, and subsurface granite was a major issue. Data prep was used to prepare surfaces where the cut went to the subgrade, and all trenches were included in the model. The surfaces were manipulated and included modifying building pad elevations and raising and lowering pipes to minimize the effect of the granite. These models were used to guide the driller, which resulted in significant savings to the contractor.
Since many projects are being phased now, Flora notes that data prep providers must analyze the entire project and all related tasks. Even though these projects are slated to be long-term endeavors, they are broken into phases for economic reasons, with each phase requiring completion prior to the next phase starting up. However, when the designers look at the job, they often view it as a single entity for earthmoving. One job Flora worked on called for 27,000 cubic yards of borrow at a cost of $225,000. According to the designer, everything was fine since the entire site balanced well. However, the borrow for this phase came from a future phase, and the erosion and siltation (E&S) plans and work had not been performed yet. This borrow was essentially inaccessible for the current phase. Data prep was done to rework the stockpiles and borrow pits to be compatible with the current phase. This model was then analyzed for the upcoming phases, and the $225,000 was saved for the client. “Dirt management is becoming more important now than ever before, and data prep is a great service provider for this type of analysis,” Flora says.
There is a famous old adage in statistics that “all models are wrong, but some are useful.”[1] Relating this to the world of 3D models for machine control, we could add that some models are more useful than others. While some 3D models offer the bare minimum of what one needs to grade a site automatically, others offer far superior value that penetrates deep into the construction technologies. Companies seeking a new business opportunity in data prep would be wise to pursue the latter. Having the capability to provide advanced analysis can pay back quickly in the high stakes world of construction.
Reference
1. Chris Anderson, “The End of Theory: The Data Deluge Makes the Scientific Method Obsolete,” Wired Magazine July 2008.
hward@carlsonsw.com
Harry O. Ward, PE, is a registered professional engineer, a state licensed contractor and certified in machine control. He is a vice president of Carlson Software and directs the Civil Engineering Division and Carlson College. He has been a member of the engineering faculty at George Mason University since 1997.
Did you enjoy this article? Click here to subscribe to the magazine.
No comments:
Post a Comment