EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
667. Enhancing Efficiency and Sustainability through Reconfigurable Delivery Systems
Invited abstract in session TC-26: Sustainability in Distribution and Transportation, stream Combinatorial Optimization.
Tuesday, 12:30-14:00Room: 012 (building: 208)
Authors (first author is the speaker)
1. | Fatima Ezzahra Achamrah
|
Management School, University of Sheffield | |
2. | Sabine Limbourg
|
HEC Liège Management School, ULiège |
Abstract
Funded by the Grantham Centre for Sustainable Futures, this research explores the development of a dynamic system for urban last-mile delivery, aimed at enhancing efficiency, resilience, and sustainability in urban logistics. Focusing on integrating real-time traffic data and predictive analytics, the study extends a previous work on a consolidation-based multi-modal delivery model, grounded in Life Cycle Assessment methodology.
Central to this research is the incorporation of real-time traffic data from platforms like Waze. This data is crucial for dynamically reconfiguring delivery routes in response to changing traffic conditions, thereby maintaining operational efficiency and resilience. Alongside, predictive analytics, utilising machine learning algorithms and demand forecasting models, plays a significant role. Focused on Sheffield area, the study also leverages data from platforms like UK Open Data for informed and automated territory reconfiguration processes. We further use Global Mapper software and clustering algorithms for dynamic adjustment of delivery territories.
The implementation strategy included a comprehensive simulation phase using AnyLogic. This phase was instrumental in modelling urban traffic and delivery scenarios, allowing for an effective comparison of the dynamic system against traditional static models. Key performance indicators such as operational costs, delivery times, and environmental impacts were analysed, providing valuable insights.
Keywords
- Logistics
- Machine Learning
- Simulation
Status: accepted
Back to the list of papers