Automatic Detection of Non-Compliant Waste

INDUSTRY : Environment
MARKET : Global
TURNOVER : €18 billion
  • Reduction in the number of shutdowns at energy-from-waste plants: the detection of a single non-compliant object avoids a potential loss of €100,000 in revenue.
  • The solution is popular with operators, improving their efficiency and precision.
  • Enhanced knowledge of operations (volumes, nature, frequency of non-compliant waste) thanks to the data collected, that can be used to prevent these items entering the waste stream.
  • Raising awareness of the potential of visual automation to help frontline workers.


As part of its waste recovery activity, the SUEZ group has set up an industrial sorting process: waste is sent to a transfer station where it is monitored remotely by operators, via screens, before being sent to a plant where it will be incinerated to produce electricity. This process requires the constant focus of the operators, who must observe up to 8 screens simultaneously to detect non-compliant items in the continuous waste stream. These are typically bulky objects, such as mattresses, metal pipes, supermarket shopping trolleys, or rubbish bins.

Play Video


When non-compliant objects are not detected by the operators, they are loaded with the rest of the waste and sent to the energy-from-waste plant. In such cases, a large object could cause a blockage, forcing an unplanned shutdown of the plant. Blockages must then be removed, which presents health and safety risks that require careful management, and repairs may be necessary to the plant before restarting the operation. Each unplanned shutdown results in a loss of production capacity, turnover, and energy revenue and represents an estimated cost of €100,000. These shutdowns can occur several times a year, despite the efforts of the teams to identify non-compliant objects in the waste. The latest innovations in the field of artificial intelligence can provide operators with more accurate monitoring tools, thus limiting the risk of unplanned shutdowns.


Last February, the SUEZ group digital team, in collaboration with Deepomatic and SUEZ R&R UK, launched a project to implement a visual automation solution at two London transfer stations, which together manage the residual waste from six London boroughs.

These two transfer stations, which supply an energy-from-waste plant, are equipped with several control screens monitored by three operators. The Deepomatic solution analyzes the images from the surveillance cameras and alerts the operators to the potential presence of non-compliant objects in the waste stream, with nearly 60 detections reported in four months. For each detection, the operator accesses images of the non-conforming item, processes an alert, and can then take necessary actions to remove the item safely.

Three steps to assist the operator in identifying non-compliant waste

  1. The operator receives an alert following the detection of non-compliant waste.
  2. He visually verifies the information and assesses the non-conformity of the waste on their dashboard.
  3. If the operator confirms the non-conformity, the waste is removed before being sent to the energy-from-waste plant.
About Suez

As an expert in water and waste management for 150 years, SUEZ is dedicated to working for the efficient and sustainable management of resources. With an active presence on five continents, SUEZ and its 90,000 employees strive to preserve our environment’s natural capital: water, soil, and air. SUEZ provides innovative and resilient solutions in water management, waste recovery, site remediation, and air treatment, optimizing resource management through “smart” cities and improving environmental and economic performance. By 2030, the Group is targeting 100% sustainable solutions, with a positive impact on our environment, health, and climate. SUEZ generated total revenue of €18.0 billion in 2019.