Motivation: #

As a civil engineer, it is common to receive thousands, if not millions, of mass points from a surveyor- this wealth of data is highly beneficial for engineers to study the land condition for an accurate execution of the earthworks.

The highly dense, closely spaced points are interconnected, creating a Triangulated Irregular Network (TIN), which provides a representation of the real world’s identical terrain.

However, users may notice that the displayed triangulations or the 3D View of the terrain is the total opposite- not all the points are used, and the terrain may appear flatter, although the survey points import is done correctly (All the points fall within the elevation range and follow the import settings).

Users may think of the above as a bug within the software, but it is, in fact, not one. The discrepancies are actually an intended behavior when very dense, closely spaced points are imported into the software, where it tries to reduce the points’ usage.

This is so-called ‘limitation’ is not permanent and can actually be resolved by applying the Use All Survey Points feature (By default, this feature has been set to No).

How does MiTS behave with closely spaced mass points? #

Closely spaced points are typically extracted from contour lines, whereby the points have almost identical X and Y coordinates with similar Z values. As an example,

In the scenario above, MiTS will detect the point as “duplication” within the dataset- the redundancy will only create complex and unnecessary elements in the triangulation network. It can’t help from translating those coordinates as flat areas, which only require fewer points for the earthwork analysis, as the cut and fill depth generally will remain constant throughout the flat stretches.

Therefore, MiTS observes the whole dataset of points based on the distance between the points and their elevation changes and only chooses dominant points– they are the points required to maintain the surface integrity, typically points that display abrupt changes in the elevations (sloping, trenches, etc.) or are located very far apart from one another. In this approach, there is no specific precision in the point selection, but the software will prioritize those points carrying significant information on the terrain’s overall shape. This reduces not only the number of points used but also the complexity of the triangulation for computational purposes while maintaining the terrain’s integrity.

What is the impact of ‘Use All Survey Points = No’ to cut and fill volumes computation? #

Though the point thinning is intended to optimize the data processing time, the major drawback of this behavior is the underestimation of the land’s exact surface.

While reducing the point usage counts, the software may unintentionally skip the ridges or valleys across the parcel of land- the triangulation lines will cut across the gap and create an illusion of a smooth surface. There is also a possibility of the software creating an artificial dip from the data interpolation during the point thinning process.

This will directly affect the cut and fill depth, followed by volumes, which is crucial in the tender and construction stages; Even with a very small discrepancy in the elevation, i.e., 10m, it will actually result in thousands of cubic meters of missing volumes from our calculation. This is something that we want to avoid, especially when money is involved in the equation!

What if the ‘Survey Point Usage Indicator’ is increased to a bigger value? Will it make the volume more accurate? #

The two features- Use All Survey Points and Survey Points Usage Indicator– are interdependent features that give users some control over the survey datasets imported.

To access the feature, you may go to

Option > Project Settings > Earthworks > Analysis

The Survey Points Usage Indicator will influence the number of points used by the software in creating a finer mesh (triangulation) between the dataset.

However, increasing the value, i.e., 1,000,000, which is twice the default value of 500,000, will only increase the points usage by a small percentage.

The increase does not really resolve the inaccuracy in the data fidelity- the number of points used in the triangulation is still half of the total point counts of the dataset.

Furthermore, there is a hard-coded limit for this feature, which is capped at 2,147,483,647 – software is not allowed to accept values beyond this point to prevent excessive memory consumption during the mesh generation.

Resolving accuracy error with ‘Use All Survey Points = Yes’ #

The inaccuracy in terrain depiction or cut and fill volumes in resolvable by setting the ‘Use All Survey Points‘ to YES in the software.

Just as its name suggests, the feature above provides granular control over the points usage for the triangulations; the software can maximize (as much as it can) the points and provide a more refined mesh for the dataset. This allows users to set how precise the triangulations should be for the earthwork computations based on their engineering judgement.

Users may not be able to observe the changes clearly with the string lines display, especially when the points are already sufficiently used to begin with, when the feature is set to No. However, we can always double-check the changes based on the Number of Nodes for the triangulations at the RHS Terrain Tab.

Why are the points involved in the mesh generation still not equal to the total point count? #

Despite activating the feature, users may notice that the points utilized for TIN do not precisely match the total point counts of the imported survey data. The discrepancies observed arise from the exclusion of the points that possess identical x,y, and z coordinates, the actual duplication, or points on the same x and y coordinates with different z levels, which is typically considered an error in the survey dataset.

In this situation, users will be warned by the software through a real-time warning message under the Verification Tab, as in the image below.

Users will be required to sort out the survey data under the Spread Input and remove the duplicated or stacked points to ensure the dataset integrity for cut and fill computation.

Comparing Terrain for ‘Use All Survey Points’: NO vs YES #

The differences in the terrain generated may not be clearly observed in the 3D View, especially when sufficient points are taken into account for the triangulation, even when the ‘Use All Survey Points’ is set to No.

However, the terrain can be compared more clearly with the cut sections, as in the diagram below.

It is observed that when the ‘Use All Survey Points’ is set to No, the triangulations created an artificial “dip” that does not exist in the raw dataset. This dip comes from the software logic of connecting points while reducing the complex triangulations for computation purposes.

Comparing by Cross Section (For road project) #

User may also use the Cross Section – All (Single View) feature to observe the impact of different options on survey points usage.

Although the highlighted disparity in cut and fill volumes might appear insignificant in the provided images, it is crucial to note that the cross-section only pertains to a small portion of the development area. However, when we consider the entire area, the cumulative volume difference will be substantial, having implications for the project.

Cut and Fill Volume: ‘No’ vs ‘Yes’ #

Results comparison: #

Discussion: #

In the industry, having discrepancies of less than 10% is already within the acceptable limit. In this particular project, enabling the Use All Survey Points increases the point count by over 50%. Still, the resulting percentage differences for the cut and fill are minimal, which is less than 1%. This is possible when the software is already employing sufficient points for the triangulations, even with default settings, and shows that point thinning behavior is highly efficient without sacrificing much on the accuracy.

However, this accuracy may be achieved if, and only if, the project scale is quite small; As the project scale increases, even a small percentage difference may lead to a higher physical volume.

Disadvantages of Using All Survey Points #

The scenario above suggests that activating the feature in your project can be an optimal solution for ensuring highly accurate earthworks computation- and we strongly endorse this approach, even for a small-scale project. However, like everything else in this world, this feature also has its drawbacks.

When the Use All Survey Points has been set to Yes, the software’s approach involves considering all possible points for slope generation and earthworks computation. This may impact the processing speed of the software. The slowdown becomes more prominent in large-scale projects, involving a large amount of survey data, particularly on machines with limited memory.

Consequently, this can hinder the progress of your design, which may be unsuitable for designers with tight deadlines and requiring efficient processing speed.

So, when is the right time to apply ‘Use All Survey Points’ to your project? #

In designing, either for small-scale or large-scale projects, we always prioritize high accuracy for the earthworks computation.

We can always choose to enable this function in all of our projects, regardless of their size or the machine we are using. However, doing so calls for patience and time investments from us.

And, it is also important to note that it is not wrong to apply point thinning, as long as the volume is acceptable. Deciding on the acceptability depends on two logics: (1) Project size and (2) Terrain Complexity. With flat or almost-flat terrains, point thinning is not significant as the volume will more or less be the same, but for terrains full of ridges or ditches, thinning of the mass points will be unsuitable due to the effects on the surface representation for earthworks computation (the inaccurate triangulations will possibly create flat areas or artificial dips due to the interpolation).

Therefore, to optimize the Use All Survey Points, we suggest a two-step configuration- During the initial design phase, the feature should be set to No (point thinning will be applied) to ensure a smooth CAD environment for designers. Then, once designers have reached the final phase, where it is required to produce tender or construction drawings, set the feature to Yes, ensuring high fidelity of the TIN in computing accurate volumes.

Conclusion #

The decision of applying the feature, or not, to a project is in the designer’s hands and should be based on your engineering knowledge. You may opt to utilize the feature, understanding that it comes with the trade-off of additional time. As the saying goes, patience is a virtue, and in this case, exercising patience will lead you to reliable output.