Posted in

Highway Construction Monitoring in Kosovo: The Pristina–Gjilan Project

Project Overview and Monitoring Need

When you drive down a new highway, it’s easy to take for granted the years of planning, construction, and countless decisions that went into its creation. The Pristina–Gjilan Highway project in Kosovo is a perfect case study in modern construction management, one that demonstrates why accurate progress tracking has become absolutely critical in infrastructure development.

The Pristina–Gjilan Highway is a significant piece of transportation infrastructure running through central Kosovo, specifically through the Municipality of Lipjan. Like most highway projects, it involves complex construction work including the highway body itself, retaining walls, drainage systems, and major structures like concrete bridges. But here’s the challenge: traditional construction monitoring methods, site visits by inspectors, manual measurements, and static photographs, simply cannot capture the full picture, especially over a project as large and spread out as a highway corridor.

This is where technology steps in. Without accurate, real-time monitoring, project managers face serious problems: delays go undetected, quality issues emerge too late, costs spiral out of control, and disputes over what was actually built versus what was designed can become expensive nightmares. For the Pristina–Gjilan project, the solution came in the form of drone-based aerial surveys combined with sophisticated software analysis, a combination that transformed how the project could be monitored and managed.

The fundamental need is simple but powerful: you need to know, with precision, whether the construction matches the design. Every drainage channel, every retaining wall, every bridge span must be verified against the original plans. When conditions on the ground force changes, those changes need to be documented accurately so the “as-built” record reflects reality, not assumptions.

Drone-Based Data Collection: The Phantom 4 RTK Advantage

The heart of the monitoring system used on the Pristina–Gjilan project is the DJI Phantom 4 RTK drone, a specifically designed surveying platform that brings centimeter-level accuracy to aerial data collection. RTK stands for Real-Time Kinematic positioning; it’s a satellite-based positioning system that corrects typical GPS errors in real time. Regular GPS satellite signals are accurate to within 2–3 meters, which is fine for finding your location on Google Maps but nowhere near precise enough for construction work. RTK corrections, delivered through a ground-based reference station or network service, narrow this down to just 1–3 centimeters of accuracy.

For the Pristina–Gjilan project, this meant that every photograph captured by the drone’s camera came with extremely precise location and altitude data embedded in it. The Phantom 4 RTK delivers:

  • A 1-inch, 20-megapixel CMOS sensor that captures detailed aerial imagery
  • A mechanical shutter that prevents blur when the drone is moving
  • Built-in RTK positioning that achieves 1 cm + 1 ppm horizontal accuracy and 1.5 cm + 1 ppm vertical accuracy
  • A TimeSync system that continuously synchronizes the flight controller, camera, and RTK module to ensure pixel-perfect positioning data

Think of it this way: imagine taking a photograph of a construction site, and every pixel in that photograph can be matched to exact coordinates on Earth within a few centimeters. That’s what the Phantom 4 RTK delivers.

DJI Phantom 4 (Source: DJI)

The Flight Mission in Kosovo

For the Pristina–Gjilan project, the research team conducted a carefully planned flight mission over a 0.515 km² segment of highway near the village of Sllovi. Here’s what the mission parameters looked like:

  • Flight altitude: 82.9 meters above ground level
  • Flight area: 0.515 km² (approximately 40 meters on each side of the highway centerline)
  • Number of flight lines: 4 routes to ensure complete coverage
  • Total photographs collected: 472 images out of 560 planned
  • Flight time: 23 minutes and 36 seconds
  • Overlap ratios: 70% horizontal overlap and 80% vertical overlap between successive photos

The high overlap is crucial. This overlap, where adjacent photos show the same features, allows specialized software to reconstruct the three-dimensional shape of the terrain and structures from the flat two-dimensional photographs.

The drone was fitted with an FC6310R camera (with an 8.8mm focal length), and the ground resolution achieved was 2.23 centimeters per pixel. This means that an object 2.23 cm across in reality would appear as a single pixel in the photographs. To put this in perspective, this resolution allows you to see individual construction details like concrete joints, cable positions, and surface irregularities.

The flight planning itself was systematic. The drone operator used the DJI GS RTK app to create the flight path, entering the highway centerline data and allowing the software to automatically generate the flight lines. Before takeoff, the drone was thoroughly checked using the DJI GO 4 app to verify firmware updates, calibration status, battery condition, and all flight parameters. This preparation is essential for consistent, accurate results.

Photogrammetric Processing: From Raw Images to Intelligent Data

Once the drone landed and the 472 photographs were safely transferred to the project team’s computer, the real work began. The photographs, while visually useful, needed to be processed to extract precise 3D information. This is where photogrammetry, the science of extracting measurements and 3D geometry from photographs, came into play. The processing was conducted using Agisoft Metashape Professional, a specialized software that uses machine learning and advanced computer vision to reconstruct 3D scenes from overlapping photographs. Here’s what happened during processing:

1. Creating the Point Cloud

The software analyzed all 472 photographs, identifying matching features across images. When the same physical object (a rock, a corner of a wall, a drainage pipe) appears in multiple overlapping photographs, the software can triangulate its exact three-dimensional position. From the Pristina–Gjilan project, the processing generated an enormous point cloud containing 265,919,213 individual 3D points, essentially a digital clone of the construction site measured in millions of tiny spots.

To appreciate the scale: imagine the entire highway segment covered with points spaced just centimeters apart, creating a complete three-dimensional representation you could visualize on a computer. This point cloud becomes the foundation for all subsequent analysis. Processing took approximately 4 hours for this project segment, which might seem long until you consider that traditional surveying of the same area would require days of fieldwork with surveyors carrying heavy equipment across difficult terrain.

2. Generating Orthophotos and Digital Elevation Models

From the point cloud, two critical products emerged:

  • Orthophotos are aerial photographs that have been corrected for lens distortion, camera tilt, and terrain variation. Think of it as creating a perfectly vertical photograph that has consistent scale throughout, like a map. The orthophoto generated for the Pristina–Gjilan segment had a resolution of 4.46 centimeters per pixel and was accurate enough to show fine details of the construction work. Unlike regular aerial photographs where distant objects appear smaller due to perspective, every square meter of the orthophoto represents the same ground area, making it suitable for precise measurements.
  • Digital Elevation Models (DEMs) are three-dimensional representations of the terrain that show elevation values across a grid. For the Pristina–Gjilan project, the DEM revealed elevations ranging from 667 meters to 761 meters above sea level across the study area. The DEM serves multiple purposes: it can be used to analyze drainage patterns, calculate cut-and-fill volumes for earthwork, assess slope stability, and visualize how the new highway fits into the existing topography. Both the orthophoto and DEM were exported in standard formats compatible with GIS (Geographic Information Systems) software, specifically ArcMap 10.5, which allowed the project team to conduct detailed spatial analysis and comparisons with the original design.

Design vs. As-Built Analysis: Where Drone Data Reveals Real Decisions

This is where the practical benefits of the monitoring system become crystal clear. Once the team had accurate orthophotos, DEMs, and the original project designs loaded into GIS software, they could overlay one on top of the other and identify, measure, and document every difference between what was planned and what was actually being built.

In construction, such differences are inevitable. Geological conditions on the ground might differ from soil surveys. Unexpected rock formations might require route adjustments. Drainage patterns might need modification due to unforeseen water flow conditions. The value of a monitoring system isn’t avoiding changes; it’s documenting them accurately so everyone knows the true “as-built” condition.

1. Changes in Drainage Channels

The Pristina–Gjilan project revealed numerous changes in the water drainage channels compared to the original design. The orthophoto and DEM clearly showed that “almost in the entire segment under consideration there has been a change in the designed state with the executed state.”

These changes came from three main causes:

  1. Initial measurement accuracy – Sometimes the original survey data used for design had lower precision than the drone-captured data
  2. Design adjustments – Engineers may have modified the original plans during construction
  3. Geological and hydrogeological conditions – The actual ground and water conditions differed from what had been predicted in the pre-construction surveys

Using GIS analysis, the project team could measure the exact location and dimensions of every drainage channel, compare it against the design specifications, and create precise records of what was actually built. This is invaluable for long-term maintenance, future maintenance crews will know exactly where drainage infrastructure is located and how it differs from older design documents.

The wells and water supply points also required analysis. The team found that while wells were generally in the same positions as designed, the water collection lines feeding them had shifted due to the changes in the canal routes. These changes, while potentially minor, could affect water flow rates, maintenance access, and long-term performance if not properly documented.

2. Retaining Walls and Gabion Structures

Perhaps the most striking finding involved the protective gabion walls designed to stabilize the highway embankment. According to the original project design, a gabion wall (a structure made of wire baskets filled with stones) was supposed to run from kilometer 12 + 655 to kilometer 12 + 915, for a total length of 300 meters.

However, when the team compared the orthophoto against the design, they discovered that at the time of the drone survey, only 172.6 meters of the wall had been completed, less than 58% of the designed length. The wall measured 14.3 meters in height at the portions that were completed.

This comparison illustrates why real-time monitoring matters. The project manager could immediately see that wall construction was behind schedule and still underway. Because the actual location and dimensions were precisely measured through the drone survey, any subsequent changes to the wall, such as extending it further or modifying its configuration, could be accurately tracked and documented.

3. Bridge Specifications Verified

For the concrete overpass bridge featured in this highway segment, the drone data allowed the team to verify and document its actual constructed dimensions:

  • Bridge length: 53.5 meters
  • Passage width (between railings): 6.0 meters
  • Total width (including side safety structures): 8.8 meters

These measurements could be extracted directly from the orthophoto and DEM through GIS analysis, providing precise documentation without waiting for a physical surveyor to climb on the structure. This approach is faster, safer, and often more accurate than traditional survey methods.

Project Management Benefits: Why This Matters Beyond Data

The technical capabilities of drone monitoring are impressive, but the real value lies in how this data improves decision-making and project management throughout the construction process.

1. Improved Quality Control

Construction quality depends on catching deviations early. When you wait until a project phase is complete to discover that work doesn’t match specifications, you face expensive options: redo the work, document the deviation and hope it doesn’t cause future problems, or engage in costly disputes with contractors.

With drone-based monitoring, project managers can conduct regular reviews, monthly, weekly, or even daily, depending on project pace, and identify issues while they can still be addressed economically. The orthophoto and DEM provide objective, measurable evidence of the current state. When combined with design documents, comparisons can be done systematically and without subjectivity.

In the Pristina–Gjilan case, the drainage changes and gabion wall status were visible and measurable as soon as the drone data was processed. Project managers could immediately assess whether these changes required corrective action or were acceptable modifications.

2. Documentation and Dispute Resolution

Construction projects often involve disputes. A contractor might claim that work conditions required a change that the owner disputes. Without precise documentation, these disputes can become expensive arbitration or court cases.

Drone-captured orthophotos and 3D models serve as objective evidence. The timestamp, GPS coordinates, and mathematical accuracy of drone-derived data make it highly credible in legal proceedings. Unlike subjective site reports or photographs taken from the ground, a drone orthophoto is georeferenced, every feature in the image can be linked to precise coordinates and elevations. For the Pristina–Gjilan project, these records will prove invaluable if any disputes arise about drainage system placement, wall configurations, or bridge specifications.

3. Archival Project Documentation

One of the most valuable but often overlooked benefits of systematic monitoring is creating a complete “archival project”, a detailed record of everything built, exactly as it was built. This becomes the foundation for operations and maintenance for decades.

After the Pristina–Gjilan Highway is completed, maintenance crews will need to know the exact location of drainage systems, the configuration of retaining walls, buried utility locations, and structural details. The drone-based monitoring system, combined with GIS analysis, creates precisely georeferenced documentation that is far superior to traditional red-line drawings (where contractors mark changes on paper copies of original plans).

This digital archive enables:

  • Rapid location of infrastructure – Need to repair a drainage system? The GIS system shows its exact location
  • Planning for rehabilitation – Comparing the as-built record with current ground conditions reveals what has changed over time
  • Efficient asset management – Future projects in the same area start with accurate baseline data
  • Training and knowledge transfer – New maintenance staff understand the infrastructure through precise, visual documentation

4. Cost and Time Efficiency

While drone surveys require upfront investment in equipment and processing, the overall project benefits are substantial. The 23-minute flight that captured data for 0.515 km² would require days of traditional surveying involving multiple surveyors, heavy equipment, and multiple site visits.

The processing time (approximately 4 hours with standard computer equipment) is a one-time investment that produces data useful for quality control, documentation, and decision-making throughout the remaining project phases and into the maintenance period. More importantly, detecting issues early, as the drone monitoring made possible with the drainage and wall configurations, prevents far more expensive problems later.

Real-World Impact

The Pristina–Gjilan case study represents more than just technical capability. It demonstrates a fundamental shift in how major infrastructure projects can be managed. Consider the practical implications:

  • For a project manager: Instead of relying on site visits and verbal reports, you have objective, measurable data about the actual construction. You can make informed decisions about schedule changes, quality issues, or design modifications based on facts, not opinions.
  • For a contractor: The precision of drone data reduces disputes. If the owner questions whether the drainage system is correctly positioned, both parties can look at the same orthophoto and measure exactly where it is. The data eliminates “he said, she said” arguments.
  • For the community: The highway is being built with proper drainage, with walls correctly positioned for safety, and with bridges meeting specifications. Future maintenance will be based on accurate knowledge of what exists, not on guesswork or deteriorated old drawings.
  • For future projects: The Pristina–Gjilan project creates a template and demonstrates the feasibility of drone-based monitoring in Kosovo. Other infrastructure projects, whether highways, railways, or bridges, can now use the same approach, improving efficiency and documentation across the entire construction sector.

Conclusion

The Pristina–Gjilan Highway project illustrates how modern technology transforms construction management. Drone-based monitoring, combined with photogrammetry and GIS analysis, provides:

  • Precision – Centimeter-level accuracy for every measurement
  • Objectivity – Data-driven documentation that eliminates subjectivity
  • Efficiency – Rapid data collection that would take surveyors weeks
  • Permanence – Digital archives that serve the project for decades
  • Confidence – Accurate as-built records that support long-term operations and maintenance

The combination of the Phantom 4 RTK drone, Agisoft Metashape Professional software, and ArcMap GIS analysis creates a workflow that is now becoming standard in sophisticated construction projects worldwide. What once required multiple surveyors with theodolites and total stations can now be accomplished by a single trained drone operator in a fraction of the time, with greater accuracy.

For Kosovo and the Pristina–Gjilan Highway specifically, this monitoring system ensures that this important piece of transportation infrastructure is built correctly, documented thoroughly, and positioned to serve the region efficiently for decades to come. As more construction projects adopt these technologies, the approach pioneered on projects like this will become the standard expectation for any major infrastructure development.

The highway is not just being built; it’s being documented for the future, one drone flight and one analysis at a time.

References

  1. Admin, & Admin. (2025, May 1). Road construction and monitoring using drones. Equinox’s Drones | Drone Company in India. https://equinoxsdrones.com/road-construction-and-monitoring-using-drones/ 
  2. Castonguay, N. (2024). Maximizing the project efficiency through comprehensive BIM coordination and GIS integration. IJERT. https://doi.org/10.5281/zenodo.18137097
  3. Digital Orthophotography and GIS. (n.d.). https://proceedings.esri.com/library/userconf/proc95/to150/p124.html
  4. Hasa, X. (2024). Drone Usage in Civil Engineering—A Case Study of the Pristina-Gjilan Highway. Engineering, 16(06), 167–180. https://doi.org/10.4236/eng.2024.166013
  5. Higgins, S. (n.d.). As-designed, as-built, as-constructed, as-is – what’s the difference? NavVis GmbH. https://www.navvis.com/blog/as-designed-as-built-as-constructed-as-is-differences
  6. Keijzers, O. (2025, December 24). Drones for construction site inspections | Acecore Technologies. Acecore Technologies. https://acecoretechnologies.com/drones-for-construction-site-inspection/
  7. Phantom 4 RTK – DJI. (n.d.). https://www.dji.com/phantom-4-rtk
  8. Sevenich, J., & Malkwitz, A. (2025). Drone-Based Construction Monitoring and BIM Deviation Analysis with Field Application and Methodological Evaluation in Bridge Construction. IOP Conference Series Earth and Environmental Science, 1555(1), 012001. https://doi.org/10.1088/1755-1315/1555/1/012001
  9. Sujit, K., Indhumathi, K., Mohan, G., Jose, N. N., Ramudu, K., Abdennaji, T. S., Cheepurupalli, N. R., Saidana, T., & Fissha, Y. (2025). Real-time construction safety monitoring using a drone based deep hybrid attention model. Scientific Reports. https://doi.org/10.1038/s41598-025-31392-5
  10. TheCivilEngineer.org. (2024, June 25). Harnessing drone technology for highway construction. TheCivilEngineer.org. https://www.thecivilengineer.org/news/harnessing-drone-technology-for-highway-construction
Share