Principal investigator: Annalisa Milella

Start: 01/10/2018

End: 23/03/2020

The E-Shelf project intends to create an eco-sustainable electronic commerce system that integrates the following components:

1. Network of food sales points aggregated in a single virtual market place;
2. Artificial vision technologies for the automatic monitoring of stocks on the shelf;
3. Mobile app for e-commerce and support for food purchasing decisions;
4. Fleet of vehicles for last mile logistics coverage, equipped with Internet of Vehicles (IoV) technologies and governed by an eco-sustainable optimization engine;
5. Social network for the loyalty of online consumers and the management of mechanisms for enhancing the reputation of both sales points and products;
6. De-verticalising middleware with high interoperability and scalability for the management of data flows.

The ISP group is involved in the task of visual shelf monitoring. The aim of this task is the design and development of a complete system based on the use of RGB-D sensors and a proper software solutions for:

• the real-time detection of out-of-stock events (OOS)
• the estimation of shelf availability (OSA)

Both goals are mandatory in warehouse and market shelf management since it is thus possible to:

• Generate alerts on OOS events
• Provide information provided for stock orders and refueling activities
• Update inventory inventory data for the e-commerce app

Out of stock detection is performed by implementing real-time planogram updates. This activity is mainly dedicated to the notification of total or partially availability of fresh products.