SelamT94/gokada-delivery-optimization

Gokada Delivery Optimization enhances Gokada's delivery efficiency in Nigeria by strategically positioning drivers. Using causal inference, machine learning, and optimization, it reduces unfulfilled requests through data-driven driver placement recommendations. # Gokada Delivery Optimization ## Overview This project focuses on optimizing the location of delivery drivers for Gokada, the largest last-mile delivery service in Nigeria. The aim is to reduce the number of unfulfilled delivery requests by leveraging causal inference, machine learning, and optimization techniques to recommend optimal driver placements. ## Project Structure - `data/`: Contains datasets used for analysis. - `notebooks/`: Jupyter notebooks for data exploration, analysis, and model development. - `src/`: Source code for data processing, modeling, and optimization. - `env/`: Virtual environment for package dependencies. - `README.md`: Project overview and instructions. ## Installation ### Prerequisites - Python 3.8 or later - pip (Python package installer) ### Setup 1. Clone the repository: ```bash git clone github.com cd gokada-delivery-optimization 2. Create a virtual environment: ```bash Copy code python -m venv env Activate the virtual environment: 3. Install dependencies: ```bash Copy code pip install -r requirements.txt ### Usage ## Data Preprocessing Preprocess the data by running the scripts in the notebooks/ directory. These notebooks include data cleaning, feature engineering, and merging datasets for analysis. ## Exploratory Data Analysis (EDA) Use the notebooks in the notebooks/ directory to perform EDA and visualize patterns in the data. ## Causal Inference Build and validate causal graphs to identify the primary causes of unfulfilled requests. The causal inference analysis can be found in the notebooks/ directory. ## Machine Learning Train machine learning models to predict delivery outcomes and optimize driver placements. Notebooks for model training and evaluation are located in the notebooks/ directory. ## Optimization Solve the driver placement problem using optimization techniques. The optimization code is located in the src/ directory. ## Results The results of t …