ChiaHsuan-Wu/WFP-Humanitarian-Supply-Chain-Optimization

A mixed-integer linear programming (MILP) pipeline designed to minimize multi-echelon humanitarian supply chain costs while satisfying non-negotiable multi-nutrient dietary requirements across 32 beneficiary camps in Ethiopia. # WFP Humanitarian Supply Chain Optimization **Operations Research & Prescriptive Analytics | Logistics Network Design & Nutrient Optimization** ### Project Overview This repository contains a prescriptive analytics solution designed to optimize the UN World Food Programme (WFP) supply chain network in Ethiopia. The model minimizes total network logistics costs—including procurement, sea freight, port handling, and inland transport—while ensuring strict adherence to nutritional requirements for 32 beneficiary camps across 11 critical nutrients. ### Tech Stack & Methodology - **Language & Environment:** Python, Jupyter Notebook - **Modeling Framework:** `CVXPY` for Linear Programming (LP) and Mixed-Integer Linear Programming (MILP). - **Constraint Modeling:** Formulated complex constraints including supplier capacities, infrastructure throughput (port/warehouse), and dietary mandates. - **Solver Backend:** SciPy/CBC for network cost minimization. ### Pipeline Architecture The project is structured as a reproducible optimization workflow: - `Supply_Chain_Optimization.ipynb`: The primary mathematical model, containing data ingestion, constraint formulation using `CVXPY`, and solver integration. - `Data OPT for DS.xlsx`: Structured dataset detailing supply chain nodes, procurement prices, transport logistics, and nutritional profiles. ### Key Optimization Insights - **Cost Efficiency:** The model optimized a flow of 65,360+ tons of relief commodities across 24 suppliers and 4 transit ports, achieving a minimized operational cost. - **Sensitivity Analysis (Duals):** Conducted shadow price analysis to identify critical bottlenecks in the network. For instance, infrastructure expansion in high-impact nodes (like Mombasa Port or Mekele Warehouse) provides direct cost-reduction paths, while expanding under-utilized nodes yields marginal returns. - **Nutritional Binding:** Identified critical nutrient thresholds (e.g., Riboflavin) that act as primary cost drivers in the n …