Abstract
In highway projects, balancing cut-fill is an important way to minimize economic payout. As a result, cut-fill calculation is a necessary task for highway designers. Traditionally, cut-fill planning is a slow, laborious process and a significant challenge in feasibility analysis. With the widespread availability of Digital Elevation Models (DEMs), automatic analysis for cut-fill projects now benefits from reliable data. This paper develops a cut-fill algorithm based on ARC/INFO GRID, utilizing a cut-fill diffusion model. Given a pre-cut-fill DEM, road coverage, engineering side-slope, and road dimensions, the algorithm calculates excavation and fill amounts and simulates the post-construction terrain. This approach is feasible within current GIS frameworks, offering significant time savings and robust design capabilities for highway engineers.
Introduction
In highway engineering, "cut-and-fill" refers to the process of excavating high ground (cutting) and filling in hollows (filling) to create a level roadbed. Balancing these volumes is critical for minimizing construction costs and environmental impact. Traditionally, engineers relied on the "average end area" method, which involves drawing 2D cross-sections at intervals along a road’s centerline and manually calculating the area between the existing ground and the proposed road. This method is labor-intensive and prone to errors, particularly in rugged terrain where ground elevation changes rapidly between sections.
This research presents an automated solution using GIS software to simulate the "post-construction" terrain. By "diffusing" road parameters—such as width, height, and side-slopes—across a digital map, the system calculates precise cut-and-fill volumes and generates a 3D model of the finished site, reducing weeks of manual work to just a few hours.
Methodology and Algorithm Design
The core of the system is an automated algorithm developed using ARC/INFO AML (Arc Macro Language). The process follows a structured workflow to ensure the final design meets engineering balance requirements.
The Diffusion Logic
The algorithm employs a "diffusion model" approach. Starting from the road centerline, the algorithm "grows" the road surface outward based on the input altitude ($Z$) and road width ($H$). Once the road width is reached, it calculates the side-slopes based on the safety slope angle ($\alpha$). To handle roads in any direction, the algorithm calculates the azimuth ($\beta$) and uses cell-stepping logic to move across the grid accurately.
Four-Way Classification of Earthwork
The algorithm categorizes earthwork into four distinct types based on the relationship between the road surface and the original terrain:
- C1/C2 (Cutting): Excavation required when the road is lower than the original ground (outside and inside road widths, respectively).
- F1/F2 (Filling): Material required when the road is higher than the original ground (outside and inside road widths, respectively).
By comparing the simulated road/slope elevation to the original terrain at every grid cell, the volume is calculated as the elevation difference multiplied by the cell area ($D^2$).
Case Study and Results
The algorithm was tested using a DEM named "TERRAIN," consisting of a 340 x 426 grid with a 1-meter resolution. The road data utilized a centerline grid at an altitude of 1750m, a road width of 30m, and a safety engineering slope of 25°.
Key Findings:
- Automation Efficiency: The authors noted a massive leap in productivity. A task that previously took five people 20 days to complete manually could be finished by one person in a few hours.
- Quantitative Accuracy: The algorithm calculated a total cut amount of 428,221 $m^3$ and a fill amount of 353,578 $m^3$.
- Generative Modeling: Unlike previous tools that were purely evaluative, this algorithm is generative. It creates a "post-cut-fill" DEM, allowing designers to visualize the environmental impact and the final topography before construction begins.
Conclusion
While this algorithm represents a significant advancement, future work is needed to account for more complex geometries, such as roadside drainage ditches or variable slope angles based on soil type. Additionally, while currently optimized for the ARC/INFO environment, the logic can be adapted to other modern GIS platforms to handle complex road networks and intersections.
Summary updated