Skip to content

akeDataAnalyst/global-wim-analytics

Repository files navigation

Global Warehousing & Inventory Analytics Dashboard

Live Demo

This project simulate realistic data flows, cleaning, KPI development, risk scoring, and interactive visualization for a decentralized humanitarian logistics network.

Project Highlights

  • Multi-country humanitarian inventory dataset generation (Ethiopia, Yemen, Bangladesh, South Sudan, Afghanistan)
  • Professional data pipeline: generation → cleaning → enrichment → analytics → dashboard
  • Composite risk scoring combining expiry pressure, stock availability, and compliance status
  • Interactive Streamlit dashboard with filters, KPIs, visualizations, data upload simulation, and audit-ready report export

Key Results

Global KPIs

Metric Value Insight
Total Items 1,500 Multi-country
Total Current Stock 2,687,965 Substantial prepositioned volume
Average Risk Score (0–100) 24.7 Moderate overall risk; significant tail of high-risk items
% Expired 31.6% Realistic expiry rate for protracted crises with access & pipeline issues
% Stockout 0.0% Strong prepositioning performance
% High Priority or Immediate Action 14.0% Items requiring urgent attention
% Non-Compliant 16.1% Compliance flags reflecting donor/audit challenges
Average Days on Hand 118 days 4-month buffer — good readiness with expiry risk

Country-Level KPI Summary

Country Items Stock Avg Risk % Expired % Near Expiry (<90d) % Non-Compliant Avg Days on Hand
Afghanistan 176 321,231 26.3 30.7% 11.9% 30.7% 118.9
Bangladesh 217 374,431 22.6 32.7% 12.4% 0.0% 119.4
Ethiopia 462 821,562 23.0 32.0% 15.8% 0.0% 119.0
South Sudan 280 508,496 25.9 31.4% 12.9% 28.6% 116.4
Yemen 365 662,245 26.4 31.0% 14.5% 29.6% 116.7

Main Insights

  • Highest combined risk in Yemen and Afghanistan (driven by non-compliance + expiry)
  • Highest near-term expiry pressure in Ethiopia (15.8% items <90 days remaining)
  • Persistent 31–33% expiry rate across all countries → distribution bottlenecks
  • No stockouts and consistent 4-month coverage → effective prepositioning strategy

Tech Stack

  • Python 3.9+
  • Streamlit (interactive dashboard)
  • Pandas (data cleaning & aggregation)
  • Plotly (interactive visualizations)
  • NumPy (numerical operations)
  • Jupyter Notebooks (phased development)

Releases

No releases published

Packages

 
 
 

Contributors