Quantity Takeoff vs. Composite Volume

[Anonymous]

I have a question about the accuracy of quantity takeoff versus computing the volume using Analyze-Volumes.  Which one is more accurate?  On the quantity take off I get around 1,000 CY less than the composite volume report on a 250,000 CY project.  Is there a way to make them match?

Thanks,

Doug

[neilyj666]

The composite volume by triangulation will always be more accurate than a sections based calculation. If you change the corridor frequency to something very small (1.0m) then the volumes will be closer but the trade of is processing time etc

[klugb]

You are talking about 0.4% difference. I doubt that your surface is that accurate. I wouldn't get too worried about making sure they match as long as they are reasonably close.

Bruce

[BrianHailey]

I think the more pertinant question is, which is less wrong?

How accurately did you model your surface? How accurately did you model your roadway?

The surface volume calculations will give you the exact volume between your two surfaces. The Quantity takeoff will give an average end area between sections.

[Anonymous]

Thank you for the responses.  Even at such a low percentage error, it would be a $25,000 difference.  I can definately see how there is a difference if the quantity takeoff uses the average end calculation method.  The problem is that we agreed to supply the surfaces to the contractor and it won't match the quantities shown on the plans.

Doug

[Anonymous]

Why not show both volumes?

Net cut (average end section method at 100 ft spacing):  ##,### CYD

Net cut (volume between triangulated surfaces):  ##,### CYD

--Jeremy

[BrianHailey]

So, you have your compaction factor and expansion factor dialed in to within 0.01%? You're good!

250,000 yard moving.

Half is cut - 125,000 yards

125,000*0.01=1,250 yards

If your compaction factor (or expansion factor) is off by that much, you will get bad results.

Earthwork quantities should not be calculated out to 5 or 6 significant digits My opinion, 1.5 significant digits (maybey 2) is good enough. If one method gives you 254,000 yards and the other gives you 246,000 yards, they both give you 250,000 yards!

[Neilw_05]

Hi Brian,

Based on your calculations I believe your expansion factor is 1%, not .01%.

I think the poster is justified in asking the question since the discrepency can be significant when multiplied by cost and shrinkage factors and it is certainly wise to cross check your numbers using alternate methods. As others have said, the TIN method is the most accurate and should be the benchmark, but there can be errors in a TIN as well, so it should not be assumed to be correct if there are significant differences between methods.

It is important to note that the cross section method does have advantages when subgrade materials need to be considred.

[klugb]

Neil,

Isn't you last sentence where this thread went sideways?

"so it should not be assumed to be correct if there are significant differences between methods."

1,000 CY out of 250,000 is not significant. The surfaces are an approximation of the ground, the sections are even more of an approximation base on the interval. If the site is a mile long and 100' wide the surfaces would have to be 12.7841' apart. If the existing ground is off by 0.05' then that's 1,000 yard off, and I would bet my house a surface that size is not that accurate.

I get tired of people in our office wasting time and money trying to model intersections down to a gnat's A$$. They will contour it down to .1' interval and not like the way the contours look and start adjusting curb grades .01' here .02' there. They need to spend  a little time in construction to understand how the design is implemented. I have a feeling they might be appalled though.

Bruce

[Kundertk]

So what you're saying, Bruce, is that you wouldn't trust the backhoe operator with Marty Feldman eyes, to tie your shoes with the bucket of his excavator? 

When I see co-workers spending too much time on these tasks, I google up an image of a scraper and point out that this is what they use to grade the site.  I agree that people get too hung up on micromanaging accuracy of Civil 3D models.  The 1000 CY discrepancy is more of a project management issue and should be discussed with the potential contractor(s) during the tender stage.  Measurement for payment is always a bit of a struggle, so the methods should be agreed upon ahead of time.

But it sure is fun to play around with the civil 3d corridor model and design surfaces to get them looking spanky.

[Neilw_05]

Don't get me wrong, I wasn't suggesting that the 1000 cy difference justified a lot of effort to reconcile it in THIS case. If a 1% discrepency adds up to a $100,000 difference then it might be a different story.

[klugb]

Neil,

I know,  I was just using your reply to keep hammering home when it's significant and when it's not.

We used to survey oil platforms in Alaska with three-wire leveling runs to .001' elevations so they could manufacture pieces in Louisiana and ship them up there. Being a few hundredths off was a big deal then!

Bruce

[Anonymous]

Hey don't knock the 0.1' contours untill you've tried them.  Sometimes there's way too much that can happen between the 1' lines.  In our area there are now organized crews of volunteers taking 2' smart levels at 16 points on every sidewalk ramp and landing, then going to the cities and regularly suing them if any of their 16 measurements exceed the ADA limits by 0.1%.  We end up often making 1"=5' scale drawings of ramps and landings to satisfy permit requirements, and to let the contractors feel confident they won't be asked to replace them at their expense.  We have to show spots at every ramp and landing corner, and slope arrows between the spots for some City permits.  Obviously we have to take the intersection curbs to that level as well.

And Contractors actually do build to those standards.  Once a contractor has been burned once by needing to replace a few hundred ramps from a season's worth of work, they start to take the detailed drawings seriously.  Of course, it has raised the cost of construction as well.

[Joe-Bouza]

And yet another oportunity to chant my mantra....

"All Contours are lies";  all it take is one look to see . or a flipped face and dont forget those designs based on 50 foot setions surveys or 20 foot in flats

[BrianHailey]

---

@Neilw wrote:

Hi Brian,

Based on your calculations I believe your expansion factor is 1%, not .01%.

---

Yes. You are right. 1%= 0.01

It's been a long day...

[BrianHailey]

---

@Neilw wrote:

Don't get me wrong, I wasn't suggesting that the 1000 cy difference justified a lot of effort to reconcile it in THIS case. If a 1% discrepency adds up to a $100,000 difference then it might be a different story.

---

Again, I don't think it is a different story. If 1% is $100,000, then the total cost is $10,000,000 and in that case, $100,000 is insignificant.

Now, when you are actually DOING the work, $100,000 can be significant but we are talking estimating here. $10,100,000 still rounds to $10,000,000.

[jmayo-EE]

Good stuff there Bruce! I agree and love it.

Keep it simple. Don't model what you don't need. Know what is significant to the project.

The OP has gotten some very good advice here from all.

[Anonymous]

Wow, this post brought out some strong feelings.  I was originally looking for an accuracy answer between the two forms of calculation.  I agree that it shouldn't be designed to the nats behind, but I wanted to know how C3D was calculating the incremental volumes in each cross-section.  The method of calculation needs to be discussed with the contractors prior to bid because the pay item is in situ cubic yards removed.  My manager's favorite saying, "You're not designing a watch".

[peterfunkautodesk]

FYI,

In 2011 there are 3 mthods to calculate the section volumes:  Average End area (with or without curve correction), the Prismoidal Formula (Simpson's Rule) or with surface to surface between stations.

Cheers,

Peter Funk

Autodesk, Inc

[blt]

The way a contractor bids a job today and the way he would bid a job years ago when we hand plotted cross sections has not changed.  We can now more accurately measure the earthwork today but are we exactly correct? No.  Therefore, in your bid documents, you specify how and what you measured and how you calculated the quantity and that you will pay on that method of calculations.  If you calculated using the avg. end method, that's how you pay, even if the contractor uses the two surfaces method and comes up with a small difference.  He has the right to show that you are incorrect or miscalculated, but only using the same method. 

Design the project to the accuracy that it needs to be built at.  You could design and build it to the 0.01' but pay for it by Lump Sum, if you specify it that way. In other words,  we do not need to design a project to the gnat's A.. for the payment side of a item quantity.

B. Thorp