teddycheru/gokada-logistic-optimization-causal-inference

Optimizing the placement of delivery drivers for Gokada, a last-mile delivery service in Nigeria, using causal inference techniques. # gokada-logistic-optimization-causal-inference Optimizing the placement of delivery drivers for Gokada, a last-mile delivery service in Nigeria, using causal inference techniques. This project aims to analyze driver locations, perform exploratory data analysis (EDA), and create causal graphs to gain insights into various operational scenarios. ## Table of Contents - Installation - Data Description - Usage - Causal Inference Analysis ## Installation To get started with this project, follow the steps below: 1. **Clone the repository:** ```bash git clone github.com cd gokada-logistic-optimization-causal-inference ``` 2. **Set up a virtual environment:** ```bash python3 -m venv myenv source env/bin/activate ``` 3. **Install the required packages:** ```bash pip install -r requirements.txt ``` 4. **Install the `causallearn` package from the cloned repository:** ```bash git clone github.com cd causal-learn pip install . ``` ## Data Description The dataset consists of two main files: 1. `driver_locations.csv`: Contains delivery requests by clients, including completed and unfulfilled requests. - Columns: - `id`: Unique identifier for the entry. - `order_id`: Unique identifier for the order. - `driver_id`: Unique identifier for the driver. - `driver_action`: Action taken by the driver (e.g., accepted, rejected). - `lat`: Latitude of the driver's location. - `lng`: Longitude of the driver's location. - `created_at`: Timestamp of the record creation. - `updated_at`: Timestamp of the record update. 2. `nb.csv`: Contains completed trip details. - Columns: - `TripID`: Unique identifier for the trip. - `TripOrigin`: Origin coordinates of the trip. - `TripDestination`: Destination coordinates of the trip. - `TripStartTime`: Start time of the trip. - `TripEndTime`: End time of the trip. ## Usage ### Preprocessing and Cleaning Data 1. **Remove `created_at` and `updated …