Update README.md

README.md

@@ -1,3 +1,67 @@
 ---
 license: mit
+task_categories:
+- reinforcement-learning
+tags:
+- operational
+size_categories:
+- 10K<n<100K
 ---
+# VRP Benchmarks
+
+This repository contains a benchmark dataset for the Vehicle Routing Problem (VRP), designed to closely mimic real-world scenarios by replicating realistic distributions of data.
+
+## Overview
+
+The Vehicle Routing Problem is a complex optimization challenge in logistics and transportation. Our benchmark provides a comprehensive dataset that researchers and practitioners can use to test and compare different VRP algorithms and solutions.
+
+## Key Features
+
+- Realistic data distributions
+- Varied problem instances
+- Comprehensive metadata
+- Easy integration with popular VRP solvers
+
+## Dataset
+
+Our dataset is generated to closely resemble real-life scenarios. It includes:
+
+- Customer locations
+- Demand quantities
+- Time windows
+- Vehicle capacities
+- Depot locations
+
+The data generation process takes into account various factors such as population density, traffic patterns, and typical business distributions to create a highly realistic benchmark.
+
+## Usage
+
+To use this benchmark:
+
+1. Clone the repository:
+   ```
+   git clone https://github.com/ahmedheakl/vrp-benchmarks.git
+   ```
+
+2. Install the required dependencies (list them here or refer to a requirements file)
+
+3. Load the dataset using your preferred VRP solver or algorithm
+
+4. Run your experiments and compare results with other solutions
+
+
+## Citation
+
+If you use this benchmark in your research, please cite it as follows:
+
+```
+[Add citation information here]
+```
+
+## License
+
+This project is licensed under the [MIT License] - see the [LICENSE.md](LICENSE.md) file for details.
+
+## Contact
+
+For questions or feedback, please open an issue in this repository or contact [Ahmed Heakl] at [ahmed.heakl@mbzuai.ac.ae].