What Is Data?

A 1.141Mb PDF of this article as it appeared in the magazine—complete with images—is available by clicking HERE

Data. Model. File. TIN. Linework. Surface. DTM. Faces. The list goes on. Whatever you call it, it is the same. It's the site specific information that is uploaded into your GPS equipped machinery and devices. It's the data that makes GPS worth the investment. If you have yet to have to deal with using it, or getting it, don't feel too bad--you are not alone. As GPS equipped machinery and devices become more accurate, more accessible and less expensive more and more companies are learning the machine control ropes for the first time. So whether you are looking into purchasing GPS machine control or have just done so, I wanted this article to answer some of the commonly asked questions that we hear on a regular basis from first time or potential users. Understanding these questions, and their answers, is critical for those who are looking at getting the most out of their GPS equipment. Because whether you are building ten miles of heavy highway, or a mine haul road or a convenience store the files you need are the same.

What is data?
To put it simply, `data' is the site specific information that is needed to build your particular site by using GPS equipped machinery. While what this data needs to contain will vary from site to site, and upon how you intend to use it, it will normally contain as a minimum the site specific linework and surface information. But it could also contain points, alignments, offsets, localization information, survey information and so on. As long as the file is complete, it allows you to move anywhere on your project and know what needs to be done at that location to bring the dirt to the design elevation. It allows you to stake out any site feature with no error. Powerful tools, indeed.

Do I need data?
If you are looking to construct your project using your GPS equipment, then yes, you need to have data. Without site specific data, your equipment will be limited to building the site `manually'--i.e. without the aid of GPS controls. In other words, a GPS equipped device without the correct data will only be able to only tell you where you are and what the elevation is there. Without correct data your equipment will need to function based on manual controls--such as construction staking in the dirt by total station, string, levels and so on. If that is solely how you intend on building your site, then no, you do not need data. If you are intending to use your GPS equipment for what it is designed for, then, yes you will need data.

Can't the engineer just provide the files?
In theory, yes. More and more engineering firms are providing both surface and linework information as part of their completed work package. Along with that service the engineer is going to require a release from to be signed before the files are provided. Read the release form carefully because it is putting all responsibility in your hands for using the files. What does this mean? This means that the engineer is not responsible for any errors in the file--whether by having incorrect information in the file or by the file missing information. This is not unreasonable for the engineer to pass this responsibility to you. Their files were not built with the intention of being used by GPS to construct their project in the field. Their files were built with the intention of designing the site sufficient enough to create a construction set of plans and ensure that each component of the design is safe and sound. By taking these files directly into the field you are forcing them to do something they were not intended to do. There is a significant amount of risk in this. Which is why I consider one of the greatest reasons to have data built is to not only provide the data, but also as a double check against the entire site for errors or omissions. The ability to work out all grade issues before ever stepping onsite is tremendously valuable.

Where can I get my model from then?
A number of places. Many companies that sell GPS will provide you with your first model or two as part of your purchase. Outside of that, you have two primary sources for getting your file built: your own company or a 3rd party. Depending on the size of your company you may have someone in your ranks who has the skill and knowledge to build models for your own use. They would need to have a full seat of the drafting and design software of their choice (typically AutoCAD; but also Terramodel, Carlson, Microstation and so on) and be comfortable with its use. They would need to be comfortable with both engineering and surveying principles. These skills and knowledge could be found in an estimator or project manager or engineer. However, whether you do not have someone who can do so, or if the qualified person does not have time, there are a number of companies that can provide this service too you. Care should be taken however in ensuring that the outsourced person/company building your file is competent, qualified and able to provide quality work. To build data requires knowledge of and experience in not only drafting/CAD work, but also as a surveyor and engineer. Because it is their job to not just create the file, but to ensure that the design is correct and plausible. When there is a 15% cross slope on a handicap parking stall they know something is amiss.

Can't we just use our take-off for the model?
As part of the bidding process, many companies will create an electronic take-off of the site. Part of this take-off is generally a surface that was used to calculate site dirt quantities by comparing it against a provided existing surface. It would not take long to come to the thought that since you already have the linework from the engineer and this surface that you have created in the take-off process that you have the components you need to put your GPS equipment to work. However, the issue with this is similar to the one with using the engineer's surface: the surface from the take-off is accurate and complete enough to get the job of a take-off done but not for fine grade. If you want a mass grade file, then your take-off surface will work fine. If it is based on contours from the engineer, its level of accuracy will be the contour interval--typically 1 or 2 feet vertical. If that level of accuracy is okay with you, then you're all set. If you need more accuracy than that, such as 0.1 foot vertical, then your take-off model will not be able to provide that. As a rule of thumb, even for mass grading, a specific model should be built for machine control.

"We have the contours from the engineer, can't we just convert those?"

This is one of the most misunderstood aspects of building data--and misused or abused. Proposed grade contours provide a specific elevation along a specific line. As soon as you move off of that line, you need another source of grade information to form a slope--usually another contour. *If* there are no grade changes between those two contours, then the slope will be correct--such as on a consistent cut/fill slope. However, as soon as there is any grade change between those two contours, your slope is no longer correct and your surface is therefore now incorrect. So whether or not you can just `convert the contours' is based on what level of accuracy you need. If you need to get the site to within 0.1 feet vertical of the design and you have a 2 foot contour interval in the plans you are going to have issues with a model based only on the contours. In addition, if your project is a roadway then the contours are entirely inadequate to represent the project correctly. You would lose high and low points, superelevation transition, intersection details, curb breaklines and so on. As a rule of thumb, proposed grade contours are a single source for building a model--not the single source. Personally, contours are the last source I use when building a model. There are much more accurate, and consistent, sources of grade information than contours.

We're ready to build the file, what is needed?
Regardless of whether someone at your company or a 3rd party is building the file, there are three consistent things that will be needed. First, the construction plans. In years past, this meant a set of plans in the mail. Nearly all projects now have electronic scans of the plans readily available. These can easily be emailed or uploaded to FTP sites to make them available to whoever is building the file. Second, some form of electronic file from the engineer--typically a CAD format is provided (.dwg or .dxf). This is where all the needed linework is coming from. It may be tempting under certain circumstances to build a file from only the electronic files and not with referencing the construction plans. This is a very risky undertaking because the electronic files are *not* signed and sealed--remember that release form? If the plans are not available you obviously have no choice. In all other cases always, always insist that the model be built from both the plans and the CAD. Third, specifications specific to your site and what is needed. Are you doing curb staking? Are you doing utility staking? Are you only responsible for mass grade? Questions like these are very important to whoever is building the data. So as a rule of thumb, you will need paper, CAD and specifications.

What should I get once the model is done?
This will vary depending on the brand of GPS you have. But the files are small and easily emailed. They will contain the surface and linework specific to your site. All that's left to be done is transfer them to your field equipment--whether directly from your computer or on a flash card--get localized and get to work.

I hope that by answering these common questions I have cleared up any confusion and made the process of getting comfortable with machine control easier. The simplest way to put it is your GPS equipment will only be as effective as your data. Bad data will give bad results in the field. Demand good data for all your projects--no matter their size. Demand data that is clear, consistent and correct. Your blade operators will thank you.

Chad Cooper (BSCE) got his start in engineering design 13 years ago with a local civil design firm, and now works on data prep projects with Take Off Professionals in Arizona.

A 1.141Mb PDF of this article as it appeared in the magazine—complete with images—is available by clicking HERE

[ Back ]

deliciousrssnewsletterlinkedinfacebooktwitter

Did you enjoy this article/ news item and want to stay informed? Click here to subscribe to the MachineControl E-Newsletter