EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
2491. Demand forecasting for Resource Planning and Dynamic Decision Making
Invited abstract in session WC-57: Forecasting for the middle mile, stream Optimization at Amazon.
Wednesday, 12:30-14:00Room: S06 (building: 101)
Authors (first author is the speaker)
| 1. | Inaki Estella Aguerri
|
| Amazon | |
| 2. | Filippo Fedeli
|
| Amazon | |
| 3. | Timoté Vaucher
|
| Amazon | |
| 4. | Giorgio Polla
|
| Amazon | |
| 5. | Thomas Ovestad
|
| Amazon |
Abstract
Every week, planning teams in Amazon need to allocate the resources (e.g., schedule trucks, plan for labor) required to deliver customer orders in a timely and cost effective manner. This planning process is based on demand forecasts, that estimate the packages to be shipped across the network. As one gets closer to the delivery date, real-time signals on the progress of the plan execution become available. This opens the door to update the forecast and refine the resource planning continuously to improve cost and increase effectiveness. However, regular plan changes need to account for uncertainty and end-to-end impact across the network to avoid plan fluctuations and local decision making. In this talk, we explore different methodologies to generate the demand forecast and its refreshes with focus on accuracy, coherency, uncertainty quantification and reduced variability, as well as its combination with different resource planning strategies for regular updating.
Keywords
- Forecasting
- Capacity Planning
- Transportation
Status: accepted
Back to the list of papers