SHOPHERO: Using loyalty card data for food waste prediction and for reducing the impact of consumer food choices

Research highlights

• Collected purchase data from 400+ loyalty cards from two Swiss supermarkets.
• Develop a food waste prediction model using loyalty card data.
• Reduce the environmental impact of consumer food choices by employing digital interventions.

Motivation

Consumer food choices can play a large role in mitigating climate change. Agricultural activities contribute greatly to global warming, accounting for approximately one quarter of global greenhouse gas emissions. In addition, roughly one quarter of the world’s produced food is either lost in supply chains or wasted by consumers, meaning that wasted food is responsible for around 6% of global greenhouse gas emissions. In developed countries, households are the largest contributors to the food waste problem.

While studies show that small dietary changes can result in a significant reduction in the environmental impact of one’s food, few studies leverage rich datasets such as digital shopping receipts or loyalty card data to assess the environmental impact of consumer food choices – and behavioral interventions targeting these choices – in the real world.

Approach

We develop a system that can automatically process consumer loyalty card data and calculates the environmental impact of their purchases. From this data we seek leverage points in digital intervention design to enable environmentally friendlier consumer food choices. In addition, we will also use the processed loyalty card data to develop a food waste prediction model.

Funding

The project is funded by the Swiss National Science Foundation.

Team

Kevin O’Sullivan (ETH Zurich), Verena Tiefenbeck, project partners: Martin Natter, Sybilla Merian (University of Zurich)