Nigeria Food Supply Chain — Demand Forecasting & Inventory Optimization

A Python-based system that forecasts weekly food commodity demand across major Nigerian markets and generates data-driven inventory recommendations.

---

Overview

Nigeria's food supply chain is shaped by seasonal agricultural cycles, festive demand spikes (Ramadan, Sallah, Christmas), and frequent supply disruptions from flooding and poor road infrastructure. This project tackles the core operational challenge faced by food commodity managers: knowing when to reorder, how much to order, and where stockout risk is highest.

The system generates a realistic synthetic dataset, runs demand forecasts, computes optimal inventory parameters, and surfaces everything through visual dashboards — all in a single notebook.

---

Features

• Demand Forecasting — 12-week forecasts per market-commodity pair using Holt-Winters Exponential Smoothing (avg. MAPE < 15%)
• Inventory Optimization — Economic Order Quantity (EOQ), safety stock, and reorder points at a 95% service level
• Reorder Alerts — 5-tier colour-coded alert system: STOCKOUT · CRITICAL · LOW · MODERATE · ADEQUATE
• Visual Dashboards — Demand trend charts, inventory heatmap, stock vs reorder point, cost breakdown, EOQ analysis
• Nigeria-specific — Encodes harmattan season, rainy season, Christmas/New Year (+60%), Ramadan, Sallah, end-of-month salary effects, and random supply shocks

---

Markets & Commodities

Markets	Commodities
Lagos, Kano, Abuja	Rice (50kg), Maize (50kg), Yam, Tomatoes
Port Harcourt, Onitsha, Ibadan	Palm Oil (25L), Garri (50kg), Beans (50kg), Groundnut Oil (20L)

---

Tech Stack

Tool	Purpose
Python	Core language
Pandas / NumPy	Data generation & manipulation
Statsmodels	Holt-Winters forecasting (ARIMA available)
Matplotlib / Seaborn	Visualizations & dashboards
Google Colab	Runtime environment

---

Getting Started

No installation needed. Open directly in Google Colab:

1. Upload nigeria_food_inventory_colab.ipynb to colab.research.google.com
2. Click Runtime → Run all
3. All outputs — dataset, forecasts, alerts, and charts — generate automatically

---

Project Structure

nigeria_food_inventory_colab.ipynb   ← Single notebook
nigeria_food_supply.csv              ← Auto-generated dataset 

---

How It Works

Generate Dataset → Exploratory Analysis → Demand Forecasting
       ↓
Inventory Optimization (EOQ + Safety Stock + ROP)
       ↓
Reorder Alerts → Visual Dashboards → Summary Report

Demand Patterns Built Into the Data

Pattern	Effect
Harmattan (Nov–Feb)	−15% demand
Rainy season (Apr–Oct)	+10% demand
Christmas / New Year	+60% demand spike
Ramadan & Sallah	+25–35% demand spike
End-of-month salary effect	+10% demand
Supply disruptions	−25% to −50% stock availability

Inventory Formulas

EOQ  = √(2 × Annual Demand × Ordering Cost / Holding Cost)
Safety Stock  = Z × σ_demand × √(lead time in weeks)    [Z = 1.645 for 95% service level]
Reorder Point = Avg demand × lead time + Safety stock

---

Sample Outputs

Reorder Alert Table

🚨 URGENT   | Lagos         | Rice (50kg bag)    | OUT OF STOCK — order 540 units
🔴 CRITICAL | Kano          | Tomatoes (basket)  | 0.8 wks left — order 1,200 units
🟡 REORDER  | Port Harcourt | Palm Oil (25L)     | Below ROP (140 < 210) — order 320 units
✅ OK       | Abuja         | Garri (50kg bag)   | 3.1 weeks of stock remaining

---

Limitations

• Dataset is synthetic — validate against real transaction data before live deployment
• Commodity prices and ordering costs are held constant; real Naira fluctuations are not modelled
• Ramadan/Sallah dates use fixed approximations (lunar calendar shifts yearly)
• Single-echelon model only (market level); does not optimise across the full supply chain