Over the past 200 years, a model backing by extracting and consuming vast quantities of finite materials and fossil fuels has been shaping the global economy. Such a linear economic model has allowed humanity to build an impressive industrial economy and reach unprecedented prosperity. At the same time, this type of economy is responsible for current environmental issues, depletion of resources, and climate change.
Per McKinsey Global Institute, AI could add $13 trillion to the global economic activity by 2030, yet some issues may constrain its application for social good. At present, the linear economic system requires changes to sustain the growth of the global middle class omitting negative environmental and social impacts.
Circular Economy, Explained
Let us start with the principal difference between the linear and circular economy models. Economic growth is not intertwining with finite resource consumption in the circular economic model. Conversely, it endeavors to eliminate waste and pollution, keep products and materials in use, and regenerate natural systems. The advantages of this approach are substantial not only for the planet but to economic growth. The circular economy can spur innovations, resolve growing environmental challenges, and create new jobs. McKinsey predicts that a net benefit of the circular economy for Europe can reach €1.8 trillion by 2030. The EU adopted the package of policies for the development of a circular economy in December 2015. Witnessing this endeavor’s positive effect, the European Commission published The Circular Economy Action Plan in March 2020, which promises more changes.
When we think of the circular economy, we often imagine waste management and recycling, such as dealing with food waste, single use-plastics, packaging, and straws. However, the circular economy is a broader concept of being sustainable, as it embraces renewable energy, design for longevity, upgrading, disassembly, water stewardship, social responsibility, and disassembly.
Artificial intelligence is supposed to play an essential role in enabling the circular economy’s systemic shift. It is to enhance and facilitate circular economy innovation across industries in three main ways:
1. Design circular products, components, and materials. It is well-known that AI can accelerate the development of new products, features, and materials fit for a circular economy thanks to the rapid ML-driven prototyping and testing.
2. Operate circular business models. AI increases product circulation by intelligent inventory management, pricing and demand prediction, and predictive maintenance.
3. Optimize circular infrastructure. AI can improve sorting and disassembling products, components remanufacturing, and recycling materials that can build the reverse logistics infrastructure required to ‘close the loop’ on products and materials.
The key idea underlying all AI applications for the circular economy is to manage resources efficiently, compliant, and sustainable. The AI technologies apply to collate, analyze and interpret complex environmental data and information to understand the issues and prioritize action. More importantly, AI can become a platform to democratize sustainability knowledge, enabling us to drive changes in our behavior that benefit the planet on all levels and scales.
From the manufacturing industry to healthcare, the scope of AI application to curb waste is endless, and the principle is relatively similar for all fields. To grasp the extent of AI applications, we can mention just several examples.
Design of New Materials
The European Space Agency deployed circulated economy models to produce and test novel alloy models in their Accelerated Metallurgy project. The circular economy principles in alloys design bring the following results: materials are non-toxic, can be reused, and can be made using additive manufacturing and processing methods to minimize waste. Accelerated Metallurgy uses AI algorithms to analyze big data to design and test alloy composition systematically.
A vital feature of the circular economy is that materials and products are not disposed of after the first use but reused multiple times, which requires optimization of the infrastructure to ensure circular product and material flows. Effective recovery of valuable materials requires homogeneous, pure flows of material and products. However, used material streams are usually far from being pristine: from kitchen waste to used computers, and these streams are mixed and heterogeneous in materials, products, and by-products, both biological and technical. AI shows how it can enable enhanced valorization of materials and products by sorting post-consumer mixed material streams through visual recognition techniques. ZenRobotics, for example, works with cameras and sensors, whose imagery input allows AI to control intelligent waste sorting robots. These robots can reach an accuracy level of 98% in sorting myriad material streams, from plastic packaging to construction waste.
Two mutually opposing trends are currently putting more pressure on agriculture, calling for immediate action. Already severely depleted soils need to provide food for an ever-growing global population, and at the same time, roughly a third of food remains never eaten. AI offers multiple opportunities to make farming smarter by using image recognition to determine fruit ripeness, food supply, demand-effective matching, and increasing food by-products valorization. Our company, for instance, uses computer vision to monitor the growth and development of plants.
Designing Healthier Food Products
AI techniques can help in reducing waste, eliminate unsafe additives, and develop regenerative grown ingredients. Recent applications include alternative egg-free products, plant-based meat, and fish to decrease dependence on natural resources. A Chilean food technology company called NotCo (The Not Company), for example, is trying to replace foods made with animal products using vegetable-based foods that taste the same. They have developed the Giuseppe artificial intelligence program that takes the molecular structure of meat and can replicate it using plant-based ingredients to create a unique flavor and texture.
AI algorithms may be able to radically improve the assessment of a product’s condition, enabling predictive maintenance and the ability to determine the secondary value of a used device more accurately. By using IoT sensors and AI-driven analytics, manufacturers and service operators can know in advance when equipment needs service. Predictive maintenance help to replace the required detail in advance. This solution predicts machine conditions leading to the failure and provides time estimation to plan and minimize downtime.
AI can be an enabler and accelerator of the global transition to the circular economy. Digital technologies are already driving a profound transformation of our economy and way of life. If such modification embraces circular economy principles, it can create value and generate more comprehensive benefits for society. However, AI production requires a clear understanding of the actual problem to solve. Moreover, the circular economy transition involves a network of trusted partners — it cannot be done by one company alone, even having the smartest AI tool.
The data generation, collection, and sharing are the implications of cooperation between all the stakeholders. It is only together with the community AI can transform our global economy and minimize waste.