Revaluing our waste through Artificial Intelligence

by | Oct 8, 2019 | AI, Use Case

waste management

 

A new deal in the global recycling market

About 300 million tonnes of plastic are produced worldwide each year, and a large part of this waste is produced in the West and sent to Southeast Asia. But receiving countries such as the Philippines, Malaysia or Thailand do not have the capacity to sort this plastic waste, and refuse to be the “dustbin of the world”. For example, Indonesia has recently returned illegally imported waste to France. Their decision to no longer import this waste is causing chaos in the global recycling market and forcing countries like France to find new solutions for managing their waste. 

 

Recovering waste through AI

France has three options for dealing with its waste, two of which – incineration and recycling – can be used to recover waste. Incineration has the advantage of producing energy: by burning one tonne of household waste, 700 kWh of electricity or up to 1,500 kWh of heat can be produced. Recycling, on the other hand, sorts the waste to transform it into a ready-to-use raw material and reintroduce it to the market. This process has many advantages, but is still underdeveloped in France. However, with the tens of millions of tonnes of waste that must be treated each year, this is a major social challenge that the State, industrialists and sorting companies must address. Waste recovery can only be done efficiently and on a large scale with the technological advances brought by Artificial Intelligence. Here are some examples:

 

Waste incineration: cost reduction and increased productivity

Today, incineration plants do not sort the waste that is sent directly from the bins to the furnaces. The furnaces burn materials that should not be incinerated (such as mattresses) blocking the supply of fuel, greasing the filters, and often causing the power plant to shut down for maintenance. The incinerator must then be cleaned so that it can be restarted. Each shutdown is expensive and polluting, and therefore harmful to both the company and the environment. Now, thanks to AI, operators can install an intelligent camera equipped with visual recognition that automatically identifies any non-compliant objects when dumping garbage containers, and sends an alert so that they can be removed before incineration. It is an effective solution to improve processes by minimizing furnace shutdowns and thus greatly reducing costs.   

 

Waste recycling: effective quality control  

You’ve probably already heard of Max AI, a robot capable of performing up to 3600 sorting gestures per hour and recognizing thousands of types of objects in order to further automate the sorting of household waste and improve recycling. However, it is still learning and making mistakes. In the current context of the global recycling market, waste quality is at the heart of national concerns, and AI can be a valuable help in this area. Since 2018 China has banned the import of 24 types of waste and refused to buy recycled plastic waste with a significant purity of 99.5%. By opting for visual recognition, sorting centres can access the necessary information in real time on the quality of recycled materials. This technology identifies the waste that is placed on the treadmill and automatically alerts an operator when a lot exceeds the compliance threshold. No need to perform quality tests by manual sampling. This reduces inventories of unsold recycled waste and improves the quality control process – a significant saving of time and money! 

 

Conclusion

The refusal of China and South-East Asia to import our plastic waste, combined with the challenges of global warming, are prompting European countries to find solutions to reuse their waste on their own soil. In France, the R&D departments of waste treatment companies increasingly consider AI as an essential tool. It enables companies to improve productivity by improving waste management and optimising sorting and recycling processes.

Share This