How many trucks between each hub pair next week? Forecast it to within one truck.
Truck-level forecasting per origin–destination pair, with lightweight optimisation on top of the existing hub-to-hub schedule – measured results from a production domestic express network.
Once a network passes roughly ten hubs and a hundred inter-hub lines, experience and rules of thumb stop working: capacity is over-provided on some corridors and missing on others, and last-minute truck orders become routine. This whitepaper lays out a validated approach that reached near-95% accuracy in predicting truck usage per corridor – and shows how to embed those forecasts in everyday weekly planning.
What is inside
The planning problem: why holiday peaks, perceived loading factors and untraceable pallet flows break manual hub-to-hub planning.
Forecast design: data foundations, back-emulation against actuals, holiday-aware features and calibrated pallets-per-truck ratios.
Proven benchmarks: ~82% exact-match on weekly truck counts, ~95% correct “add-truck” recommendations, 87%+ of weekly hub-pair decisions automatable.
Five practical takeaways for logistics professionals moving to truck-level forecasting.
Get the whitepaper
Enter your details and we will email you the PDF right away.
By submitting, you agree to our privacy policy. We will email the PDF and occasional, relevant updates – unsubscribe anytime.
Prefer to see it on your data?
Book a 30-minute demo and we will walk your network – using a slice of your own data.