Image analysis with AI to facilitate furniture renovation decisions and increase circularity
01 March 2023, 08:25
RISE is leading a research project that builds a decision support system for actors in the furniture industry. The aim is to use AI to create a tool that can make cost assessments, classify products and calculate environmental impact in order to make quick and correct decisions about furniture renovation.
- Refurbishment is an important part of the emerging circular economy, but to scale up refurbishment profitably, companies need to be able to make complex decisions faster and more efficiently. The digital tools we want to develop will enable fast and accurate decisions, which is a prerequisite for circular business models to take off, says project leader Robert Boyer from the Sustainable Business unit at RISE.
Decision-making in the circular economy can be challenging due to reverse logistics and lack of readily available data. The project run by RISE is called Digital Decision Support for Refurbishment (2DSR) and will run until 2025. It is funded by Vinnova and involves furniture players from across the value chain; OOAKI Living, Kinnarps, GoZero Sweden AB (formerly Stafe Group AB) and Eskilstuna Omlastningscentral. OOAKI Living is a company that, among other things, gives new life to sofas by renovating them with new upholstery.
- We want to make it easy for our customers to make a conscious, environmentally friendly choice. Our goal is that consumers should first consider keeping their sofa by renewing it with a new upholstery, and only secondly think about buying a new sofa. We believe that with the help of AI technology, the furniture industry become more sustainable, environmentally conscious and circular, says OOAKI Living's founder Andreea Björnsson.
The project will use machine learning to develop an image-based algorithm that supports refurbishment decisions by quickly identifying furniture models, thus speeding up the identification of spare parts and calculations that traditionally require costly expert assessments and/or extensive resources.
- The digital decision support we are developing will analyse thousands of digital images and existing digital models to train a computer program to recognise and evaluate furniture, says Robert Boyer, RISE.
Robert Boyer, RISE
firstname.lastname@example.org, 010-228 40 66