Using Deep Learning to Forecast Demand for Thousands of Grocery Items


Ocado Group provides end-to-end online grocery ecommerce, fulfilment and logistics solutions to some of the world’s largest grocery retailers. For the partners that Ocado works with, having the right stock available at the right time and in the right place is critical for operating their online grocery businesses successfully. To achieve this, Ocado’s technology teams have developed highly accurate demand forecasting models.

Forecasting product demand for online grocery accurately is challenging: shopper behaviour is constantly changing, extremely large ranges of mostly low selling products make the demand noisy, and demand forecasts are required from the moment an online retailer launches, with no past sales history.

Sam Blake, Ocado Technology’s supply chain data science lead, will dive into how the company uses deep learning to overcome these challenges and build robust demand forecasting models that ensure customers can order and receive the products they want with as little stock wasted as possible.

Everyday the company’s forecasting models generate predictions for grocery retailers across four continents, predicting daily demand months into the future for every individual product stocked by each fulfilment site. This equates to hundreds of millions of forecasts every day.

Sam will lift the lid on the deep learning models used by Ocado Technology and how these have been adapted for the challenges faced in online grocery, showcasing the positive results achieved by the retailers who have adopted these forecasting solutions including 50% improvement in accuracy, drastically reduced waste, automation of replenishment decisions and big financial savings.

Join this session to get a glimpse into a real life example of deep learning in production and how it is having an impressive impact in the ecommerce space.


Sam leads the supply chain machine learning team at Ocado Technology, responsible for the demand forecasting and replenishment optimisation algorithms used by Ocado’s international partners.

Sam holds a DPhil in Condensed Matter Physics from the University of Oxford and volunteers as an ambassador for DataKindUK. Prior to joining Ocado he spent a number of years working in AI startups in the Netherlands.

Open Data Science




Open Data Science
One Broadway
Cambridge, MA 02142

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from - Youtube
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google