A first machine-learning pilot for retail demand forecasting.
Energy & fuel retail · National retail network
- Challenge
- Forecasting across a national retail network relied on manual effort and couldn't keep up with demand variability.
- What I did
- Framed the use case with the business, set up the proof of concept with data and IT teams, defined success criteria and built the path from pilot to network-wide rollout.
- Outcome
- The organisation's first ML initiative, proving value at network scale and creating a template for future AI use cases.