Applying Artificial Intelligence to Circular Economies

The climate crisis is one of the greatest challenges of our time. In our quest to mitigate the consequences of climate change, we need far-reaching change processes. This poses special tasks for companies because it involves more than simply switching to green energy. It also means developing truly circular products with multiple life-cycles in the market. It is about designing products so they can be maintained and updated to extend their lifetime as much as possible. When products reach the end of their life it makes sense to remanufacture or recycle them so that materials can be reused. The purpose of the circular economy is therefore to achieve a closed resource cycle. Such an approach minimises the negative environmental impacts of products and contributes to sustainable development by transitioning to a resilient economy.  The circular economy concept is an approach that can help achieve the sustainable development goal of "responsible production and consumption" (SDG 12) and lead to driving innovation (SDG 9) in the respective fields.

The graphic below shows the cycles of a circular economy.

AI meets circular economy 

One of the challenges the lighting industry faces in establishing a circular economy is how to handle electronic components at the end of a luminaire’s life. In the best case, the electronics will be recycled. Luminaires are highly manufactured products, so their recycling process involves high costs. In many cases, however, fully operational parts are disposed of with no chance of being reused. As the stakeholders in the value chain do not have information about the components’ status and remaining lifetime, the necessary structures for harvesting functional components do not exist.

At Tridonic, we want our products to have the longest possible lifetime, which is why we utilise machine learning. This technology enables us to generate additional data about our products. This new information can be used to optimise the design of new luminaires and help manufacturers and users alike in deciding when to reuse, remanufacture or recycle their products. By leveraging our expertise, they can make more sustainable choices and the environmental impact of lighting can be significantly reduced.

How does machine learning help? Tridonic uses learning algorithms to analyse large amounts of data on the performance of lighting components, such as the type and age of the component, the operating conditions, and any previous failures. By analysing this data, the algorithm can learn to identify patterns and interdependencies indicative of a component’s lifetime.

Machine learning can make a difference
At the many stages of the technical cycle shown in the graphic above, machine learning can make a difference. As a subset of artificial intelligence, it involves using algorithms and statistical models to enable a system to improve its performance on specific tasks based on previous learnings. Lighting components produce valuable data throughout their life-cycle, including operational data from connected lighting installations. By applying artificial intelligence to these datasets, we can gain valuable insights into the performance of lighting components.

Luminaire maintenance can become more efficient when product-specific lifetime calculations for lighting components are available. Manufacturers and consumers will then be in a position to replace only those lighting components that need replacing, thereby reducing waste and keeping luminaire in operation. This is also of benefit to companies in charge of maintenance, as they can optimise their planning based on forecasts.

One of the key benefits of Tridonic’s approach is the potential to change the way decisions are made at the end of a product’s life. Armed with the generated data, luminaire manufacturers can make more sustainable choices with the confidence that their products will still offer the quality and performance their customers expect after the luminaire’s initial usage stage. The natural consequence is that the environmental impact of lighting will reduce as the rate of reuse and remanufacture of electrical components increases.

In conclusion, by leveraging the power of machine learning, manufacturers and consumers can make more sustainable choices and reduce the environmental impact of lighting, resulting in significant savings in resources and emissions.

Luis Javier Carracedo Cordovilla uses his long-term experience in digitalisation, Industry 4.0, and IoT to explore, together with his business innovation team, how the lighting infrastructure can positively impact our world by creating new ways of experiencing light.

Tridonic is a world-leading supplier of lighting technology, supporting its customers with intelligent hardware and software, and offering the highest levels of quality, reliability and energy savings. As a global driver of innovation in the field of lighting-based network technology, Tridonic develops scalable, future-oriented solutions that give rise to new business models.

Twitter | LinkedIn | Youtube

Contact form: https://www.tridonic.com/com/en/contact-form.asp

Der Gastbeitrag wurde im Rahmen des Projektes DIGI FOR SDG veröffentlicht.