A project by TUM and Hochschule München.

Phoenix AI

A marketplace for big corporations to procure large amounts of recycled materials.

Challenge

Resource Take-Back for increase in City of Munich resource efficiency

The challenge is to ensure transparent and consistent resource flows to provide recycled material of certified quality and in large quantities to meet BMW's needs and help the city of Munich achieve its recycling targets.

Team

Katharina Grünwald, Sebastian Wagner, Harsh Mulrav, Eduard Krasnov

About the prototype

A design thinking methodology was applied to address supply chain inefficiencies. The solution involves two main parts: predictive analysis using machine learning to ensure a reliable supply of recycled materials and process optimization for transparency and reliability. A machine learning model, trained on ERP data from waste collection centers, predicts material availability accurately. Additionally, a decision-making model using AHP and TOPSIS methods rates suppliers, and bundling small quantities of materials increases economic viability. The innovative business model leverages a modern, integrated platform serving recyclers and manufacturers, developed using the business model canvas. This platform increases recyclers' revenue by enabling the recycling of previously non-profitable waste through efficient bundling. For manufacturers, it guarantees a steady supply of certified recycled materials, aiding in meeting sustainability goals and securing raw materials at competitive prices. The combined software solutions ensure a smooth, transparent process of supplying and procuring recycled materials, benefiting all stakeholders.

Outputs

  • pdf
    Project report

    Further information about the progress, milestones, and roadblocks.

  • Credits: Photo by students