Optimising Electronic Waste Collection and Recycling in Brussels

Organisation type

Public administration at local level

Use case area

Environment

Geographic scope

Belgium

Challenge:

The growing volume of electronic waste presents a significant environmental challenge. Many electronical appliances and light bulbs contain hazardous materials such as mercury and rare metals like gold, silver, and cobalt. Improper disposal of these items can lead to the release of harmful substances into the environment. Efficiently sorting and recycling these materials is critical but often hampered by manual processes and lack of accurate data. The challenge is to develop a system that can optimise the recycling process, reduce waste, and promote circular economy.

Approach:

The solution employs machine learning to enhance the city’s operations in collecting and recycling old electronic devices. When a truck filled with discarded devices arrives at an interim station, each device is placed on a weighing platform. Six different cameras take photos of the devices, and this data is uploaded to a cloud-based database. A machine learning model then analyses the data to predict the type of product.
Initially, the algorithm could only categorise devices into broad groups like large household appliances, gardening equipment, and IT equipment. Since 2020, the algorithm has been refined to recognise specific devices, such as hedge trimmers, weed burners, and blenders. By leveraging machine learning techniques, it is possible to analyse a vast dataset comprising information on electrical appliances and light bulbs delivered for recycling. By training models on these data, the system can predict the most efficient recycling paths for different items, identifying valuable materials for recovery, and optimising logistics for collection and processing.
Collaborating with local recycling centres and authorities, the project aims to create a streamlined and effective recycling system that is continuously fed with relevant data. This detailed information is crucial for recyclers, as it helps them identify the types of valuable materials they can recover and resell from the devices. Additionally, knowing the specific devices allows recyclers to better estimate the effort and cost involved in the recycling process.

Outcomes:

The implementation of this initiative is expected to improve recycling rates and reduce the volume of waste sent to landfills. In 2023, 95% of collected e-waste was repurposed or recycled, with nearly 7 million kg of appliances reused. By efficiently recovering valuable raw materials, the project will contribute to resource conservation and support local economy. Additionally, the reduction in hazardous waste will have a positive impact on public health and the environment, fostering a more sustainable community.

Data Sources:

In this use case, image data on electrical appliances delivered for recycling serves as a foundational element. Additionally, data on used lamps and bulbs is leveraged to enhance accuracy. Furthermore, historical recycling data provides a comprehensive view, allowing for more informed predictions and decisions.

Additional Information:

Point of Contact:

cellule [dot] webbrucity [dot] be (cellule[dot]web[at]brucity[dot]be)