Related Concept Videos
Distributed Loads: Problem Solving
Short-distance Transport of Resources
Distribution Reliability and Automation
Distributed Loads
For example, consider a bookshelf filled with books stacked vertically adjacent to each other. The weight of the books is evenly distributed over the length of the shelf. As a result, the pressure at different locations on the surface of the...
Applications of GIS: Disaster Management and Emergency Response
GIS Software, Hardware, and Sources of GIS Data
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
Agentic LLM-based robotic systems for real-world applications: a review on their agenticness and ethics.
Decomposing user-defined tasks in a reinforcement learning setup using TextWorld.
Related Experiment Video
Updated: Sep 17, 2025

Deep Neural Networks for Image-Based Dietary Assessment
Published on: March 13, 2021
Delivering data: A real-world dataset for last-mile delivery optimization.
Anna Vrani1, Savvas D Apostolidis1, Athanasios Ch Kapoutsis1
1Information Technologies Institute, The Centre for Research and Technology, Hellas, Thessaloniki 57001, Greece.
This dataset provides structured distance and time matrices for optimizing Vehicle Routing Problems (VRP) in logistics. It enables realistic modeling of delivery operations using real-world traffic data.
Area of Science:
- Operations Research
- Logistics and Supply Chain Management
- Data Science
Background:
- Vehicle Routing Problems (VRP) are critical in logistics for optimizing delivery routes.
- Real-world VRP optimization requires accurate travel time and distance data, accounting for traffic variability.
- Existing datasets often lack comprehensive, real-world traffic-influenced travel time information.
Purpose of the Study:
- To present a novel dataset structured for Vehicle Routing Problem (VRP) optimization.
- To provide realistic travel time and distance matrices derived from actual logistics operations.
- To facilitate the benchmarking and development of VRP algorithms using industry-relevant data.
Main Methods:
- Data acquired from a Third-Party Logistics (3PL) provider's order management system.
- Distance and time matrices generated using an API incorporating historical and real-time traffic data.
- Four matrices created: one for distance, three for travel time (optimistic, pessimistic, most likely traffic scenarios).
Main Results:
- A dataset comprising nine days of structured distance and time matrices for 60-85 daily delivery stops.
- Matrices offer a complete view of travel times and distances between all locations, crucial for VRP.
- Data reflects realistic travel conditions by considering traffic congestion variability.
Conclusions:
- The dataset serves as a valuable benchmark for evaluating and comparing VRP algorithms.
- Enables statistical analysis and Monte Carlo simulations for modeling routing uncertainty.
- Supports optimization of real-world delivery operations and addresses industry logistics challenges.