789. Integrated forecasting and optimization in multi-echelon logistics
Invited abstract in session TB-38: Forecasting, prediction and optimization 2, stream Data Science meets Optimization.
Tuesday, 10:30-12:00Room: Michael Sadler LG19
Authors (first author is the speaker)
| 1. | Vittorio Maniezzo
|
| dept. Computer Science, University of Bologna | |
| 2. | Livio Fenga
|
| University of Exeter |
Abstract
The increasing complexity of Large-Scale Retail Trade (LSRT) supply chains calls for efficient multi-echelon logistics optimization. The efficiency of these networks typically depends also on optimized inventory management, transportation, and demand forecasting and can benefit from predictive and prescriptive analytics for cost-effective decision making. This paper presents a novel approach that integrates predictive analytics with optimization to improve inventory allocation through customer demand forecasting. A case study of LSRT logistics is described, requiring demand forecasting, optimization of distribution center (DC) sizing and retailer allocation. The approach uses a bootstrap-based forecasting model for demand data that is seamlessly integrated with a deterministic equivalent optimization component. The analysis is extended to include integer split assignments for scenario validation. The study demonstrates the effectiveness of this combined forecast-optimization pipeline through computational analysis. The results are validated with real-world data and demonstrate practical applicability and superior accuracy and efficiency, thanks to a generally applicable approach not limited to logistics decision making.
Keywords
- Forecasting
- Stochastic Optimization
- Logistics
Status: accepted
Back to the list of papers