United KingdomRetail and e-commerce2023 – ongoingJohn Lewis Partnership
Production machine learning and bakery optimisation
Dataiku attributed £40M in value to John Lewis Partnership's AI work
Problem
Waitrose bakers relied on experience and manual spreadsheets to decide daily baking quantities. John Lewis Partnership also needed a repeatable way to put machine-learning models into production.
Approach
The partnership worked with Deloitte and Dataiku on production model workflows. Its data science team built a bakery model using historical sales and additional datasets. Dataiku ran the model workflow, while Tableau presented daily bake recommendations to store staff.
Results
Takeaway
The bakery application put predictions into daily store decisions. The £40M vendor claim covered the wider programme and was not a bakery-only saving.
Categories
- Size
- Large enterprise
- Industry
- Retail and e-commerce
- Business function
- Production and qualityIT and R&D
- Innovation type
- Process innovation
- AI technology
- Predictive analytics and MLOptimization
- Adoption
- Partner co-development
- Evidence
- Vendor case study
Sources
Source summaries
Dataiku attributed £40M in delivered value to John Lewis Partnership's production models. It described the MLOps programme with Deloitte and the bakery forecasting application.
DataIQ described Waitrose's bakery forecasting model and its daily recommendations. The award account reported reduced spreadsheet work and commercial gains.
Full details are in the original sources.