Drones and AI to Improve Construction Waste Management Systems

Dubai is the epitome of construction ambition where the cityscape often changes from quarter to quarter. That’s how fast-paced the construction ambitions of this magical city are! That being said, it’s no wonder that one of the city’s leading property builders approached us to assist them in their Site Clearance process.

With the assistance of drone-powered solutions, construction waste management is now simpler and more efficient. Moreover, when the data analysis is automated, it further escalates the advantages of drone-powered solutions. And that’s how the team FEDS approached this mission – Drone-powered solutions for quick, safe and efficient data collection with AI-assisted data analysis.

You can find several inspiring examples of how Drone Mapping is helping cleaning up the world. Well, this case study talks about how we helped a property owner of the region with managing construction wastes left behind their plots. The data we provided enabled them to precisely locate waste materials and remove them with immediate effect before initiating their prospective projects. Drone-powered solutions are economical, practical, efficient, and accurate alternative to the existing workflow. Let’s see how.

The Mission

A leading property builder in the UAE approached us to assist them in their site clearance efforts. A Site Clearance is an effective approach to remove any unwanted materials and speed up the construction process for prospective projects. However, clearing a plot of land isn’t that simple but rather a challenging liability at times. For instance, suppose we have to remove the construction wastes scattered across acres of land. In this scenario, if we were to go with the typical method on foot – locating wastes, picking up and removing waste materials, it would have been a heavily labour-intensive, time-consuming and extravagant process, further delaying all subsequent tasks. But, thanks to drone-powered solutions!

The property builders wanted us to immediately help them locate construction wastes left behind in their underdeveloped plots. We took up the challenge- team FEDS conducted a Drone Survey to find construction waste left behind across 22 abandoned plots within a residential zone of Dubai. “To speedup the efficiency of this time-sensitivity of the project without losing the accuracy, we decided to deploy multiple drones and payloads from our fleet,” says Nikin Mohan James, Operations Manager, FEDS. Using three sets of powerful DJI M300 RTK paired with DJI Zenmuse P1. An effective and precise map of the 3 sqkm area was produced in just five hours. With photos and videos captured by drones flying seamlessly over the terrain, managers analysed the topography of their plots. The live-streaming of the inspection also enabled the client with quick insights into their plots.

AI-powered Data Processing

The traditional methods of data processing would have given us accurate results, but the amount of time and effort involved in the process cannot be neglected. That is how the team decided to automate the data analysis with the help of Artificial Intelligence. What’s the difference? Data Analysis as accurate as the manual data processing but 90% faster! Our GIS engineer Sreya Madhavan trained an AI detector to identify the construction waste piled up on the plots. She trained the detector by feeding it with potential construction waste samples from the data captured by drones. The most challenging aspect was the mixture of heterogeneous construction waste; it would have been much simpler if we were to train an AI detector to identify specific construction waste, for instance, tyres or manholes. However, despite the limited sample size and the mixed-up construction trash, the FEDS Data Experts were able to train the AI Detector in under 2 hours.

Construction Waste
Here are a few examples from the AI library - the sample images that were used to train the AI detector.
AI-spotted Construction Waste
A preview of how AI located construction wastes in the plot.

Mission Deliverables

The AI detector located the construction wastes along with the location and other details under an impressive 15 minutes, which would have otherwise taken weeks to months. The Project deliverables included Orthomosaic maps, RGB and Thermal videos, Photos, report containing the location of construction wastes and finally, a full-fledged comprehensive mission report. The Orthomosaic map contains coordinates in each pixel, enabling us to pinpoint where exactly the construction wastes were piled up. This can be utilised to locate and remove the wastes without further delay.

Clustered Construction Waste
The wastes identified for each plot were clustered together to determine critical areas. The colour coding goes from red to blue; red denotes small clusters, while blue denotes large clusters.

Drone-powered & AI Enabled solutions with FEDS


The fact that there are still some professionals out there sifting through thousands of photos of plots, to look for tiny construction wastes is really bothering; it is high time that we switch to smart solutions for better and faster results. I would like to reiterate how automated data analytics further escalates the advantages of drone-powered solutions. And that’s what we do; Drones for fastest data collection and AI-assisted data analysis!

Are you a professional in the Construction or Waste Management sector? I’m glad that you are here right now. Using Artificial Intelligence in conjunction with the ever-evolving Drone Technology, we empower you with the best solutions. Our solutions are economical, practical, efficient, and accurate alternative to the existing workflow. If you’d like to further explore the impact of drone-powered solutions in your operations, do not hesitate to reach out to us.

About the author

Maria Thomas

Maria Thomas