Data Science & AI for Circular Economy.pdf

Sustainability

1. Data Science & AI
for Circular
Economy
Parvathy Krishnan Krishnakumari
Chief Technology Officer, Analytics for a Better World Institute
Senior Data Scientist (Consultant) for UNDP and World Bank
[email protected]
LinkedIn - parvathykrishnank

2. Technology is crucial to realising this vision at scale
The circular economy system

3. Data science and Artificial Intelligence as an enabler
The circular economy system

4. Data science and Artificial Intelligence as an enabler
The circular economy system

5. Data science and Artificial Intelligence as an enabler
The circular economy system

6. AI-Driven Material Innovation for Circular Economy
Funded by the European Space
Agency, the project ‘Accelerated
Metallurgy’ conducted research on the
rapid and systematic development,
production, and testing of novel alloy
combinations.
Accelerated Metallurgy uses AI
algorithms to systematically analyse
huge amounts of data on existing
materials and their properties to
design and test new alloy formulations.
By capturing details of the chemical,
physical, and mechanical properties of
these unexplored alloys, the algorithms
can map key trends in structure,
process, and properties to improve
alloy design using rapid feedback
loops.
The project aimed to develop new
metals with the same performance in
a more efficient way. Alloys designed
with circular economy principles in
mind are non-toxic, are designed to be
used and reused, have longer use
periods, and could be made using
additive manufacturing and
processing methods that minimise
waste. Additionally, improved material
properties can implicitly reduce
resource use through enhanced
product performance.
01 02
03 04

7. Zen Robotics - Revolutionizing Recycling with AI
Innovation at Work: ZenRobotics
employs advanced AI-driven
technology to address the sorting
challenge, using cameras and sensors
to distinguish and separate materials
with precision.
Visual Recognition: The AI system's
ability to recognize various materials
through visual input is a game-
changer in waste management.

8. Power of Geospatial Data
Circular Economy and GIS
Resource Management: Tracking and managing natural resources for sustainable
usage.
Supply Chain Optimization: Monitoring and optimizing logistics to reduce emissions.
Agricultural Practices: Enhancing soil health and crop rotation for sustainable
farming.
Urban Planning: Assisting in the design of cities that support the circular economy
through waste reduction and resource efficiency.
Disaster Response: Enabling efficient debris management and resource allocation
post-natural disasters.
Environmental Monitoring: Observing changes in ecosystems to ensure
environmental restoration and conservation efforts align with circular principles.

9. Emissions monitoring for landfills and dumpsites

10. Power of Natural Language
Circular Economy and NLP
Text Mining for Resource Efficiency: Analyze industry reports to discover materials
that reduce environmental impact.
Sentiment Analysis for Product Design: Use customer feedback to create products
that meet circular economy standards.
Regulatory Compliance Tracking: Evaluate legislation and policy documents to
ensure circular practices.
Market Trend Insights: Identify sustainability trends from academic and market
research for strategic planning.
Supply Chain Transparency: Scrutinize supplier documents to find opportunities for
recycling and reuse.
Waste Stream Analysis: Use textual data to optimize waste sorting and recycling
processes.

11. Automatic Analysis of Textual CSR Data

12. Conclusion
To harness AI’s full potential:
We must identify opportunities across
sectors and set the stage for innovation.
It's essential to raise awareness about AI's
role in circularity.
Stakeholders can draw inspiration from case
studies to explore AI applications in their
fields.
Experts in circular economy and AI must
collaborate to expedite their goals.
Investors are encouraged to fund AI ventures
that support circularity.

13. Conclusion
For effective AI applications:
We need open, collaborative data sharing
and high-quality data labeling.
The success of AI in the circular economy
depends on cross-sector collaboration.
Above all, the ethical development of AI is non-
negotiable:
Fairness, privacy, and security must be
integral to AI systems.