Optimizing supply chain operations of SMEs in Ghana using machine learning for demand forecasting and inventory management.
# supply-chain-ml-smes-ghana
# Optimizing Supply Chain Operations of SMEs in Ghana Using Machine Learning
## 📌 Project Overview
This project applies machine learning techniques to improve **demand forecasting**
and **inventory management** for Small and Medium-sized Enterprises (SMEs) in Ghana.
The study is based on **primary data collected through structured questionnaires**
administered to SME owners and managers, capturing inventory practices, demand patterns,
and operational challenges.
The goal is to demonstrate how data-driven decision-making can help SMEs reduce
stockouts, minimize overstocking, and improve supply chain efficiency.
---
## 🎯 Objectives
- Analyze demand and inventory patterns of SMEs in Ghana
- Apply machine learning models for demand forecasting
- Identify key factors contributing to stockouts and excess inventory
- Provide actionable, data-driven recommendations for SMEs
---
## 🗂️ Data Description
- **Source**: Primary data collected via questionnaires
- **Respondents**: SME owners and managers in Ghana
- **Format**: Excel (.xlsx)
- **Key Variables**:
- Business characteristics
- Inventory replenishment practices
- Demand variability indicators
- Stock availability and frequency of stockouts
> ⚠️ Note: The dataset has been anonymized and is used strictly for academic and analytical purposes.
---
## 🛠️ Tools & Technologies
- Python
- Pandas, NumPy
- Scikit-learn
- Matplotlib, Seaborn
- Jupyter Notebook
---
## 📊 Methodology
1. Data collection through questionnaires
2. Data cleaning and preprocessing
3. Exploratory Data Analysis (EDA)
4. Feature engineering
5. Machine learning model development
6. Model evaluation and interpretation
7. Business insights and recommendations
---
## 🔍 Key Insights
- SMEs with high demand variability experience more frequent stockouts
- Poor inventory planning is strongly associated with excess inventory
- Machine learning-based forecasting improves demand estimation accuracy
- Data-driven inventory de …