Demand forecasting and ETA prediction for a last-mile operator

Last-mile operator · Logistics

Unified order, fleet and weather data feed forecasting and ETA models used by planners and customers every day.

Challenge

Planners set capacity by hand from last week’s numbers. Peaks caused missed deliveries, quiet days wasted fleet, and customers received ETAs that were often wrong by hours.

Solution

We consolidated orders, telemetry and external signals in Snowflake, built probabilistic demand forecasts per zone and hour, and an ETA model retrained daily on live delivery data. Results are served through an operations control tower and the customer app.

Forecast horizon of 14 days with confidence intervals
ETA predictions refreshed every 5 minutes
Scenario planning for peak seasons

Ready to put your data to work?

Tell us about your challenge. We will come back with a concrete first step within two business days.

Get in touch