ReWaCC - Remote Waste Characterization and Quality Control

Waste containers or skips are everywhere in our urban landscape. Today, the waste companies often don't even know exactly how many containers they have, and much less where they are and how full they are. This means they are picked up only after someone calls the waste company and letting them know that it is full. Sensorita has developed a container-mounted sensor giving data on where the containers are and how full they are. This project aims to use the sensor data along with other data sources to also measure what kind of waste is in the containers. With this information, waste companies knows what waste is coming their way, enabling them to plan their production better, thus bettering the recycling process.

Waste is one of the most significant challenges of our time, and although the EU has implemented legislations to foster the transition to a circular economy, the lack of digitalization remains a major obstacle. Specifically, construction and industrial waste hold significant untapped value as secondary materials. This project aims to provide digital twins of waste containers with near-live waste characteristics, transforming containers into a decentralized material warehouse. By utilizing sensors and data analysis, stakeholders gain valuable insights for operational and strategic decision-making, revolutionizing waste management practice. The project's positioning lies in its unique remote waste quality monitoring capabilities, offering a scalable solution based on retrofittable sensors for existing waste containers. To unlock the potential of waste characterization, the project leverages radar and machine learning. By extracting rich information from radar measurements, waste characteristics can be inferred, with additional data sources like container positions and weather data further enhancing the accuracy. A comprehensive picture of the waste can be constructed, providing insights into waste composition and quality. The solution facilitates better recycling and increases the opportunity for secondary materials to replace virgin materials. R&D challenges include quantifying downstream effects, identifying patterns in non-radar data sources, and constructing machine learning algorithms that infer waste characteristics. These challenges require an interdisciplinary approach where understanding of waste generation behavior, radar theory, machine learning, and knowledge about waste based processing and production must be combined. The project is a collaboration along the waste value chain, including Veidekke and Ragn-Sells, and is supported by R&D suppliers with sustainability, machine learning and radar expertise; NORSUS, Norsk Regnesentral and Acconeer.