YvesNDIKURIYO-2022/En-NSGA-II-4-LC-CTRP

Enhanced NSGA-II framework for low-carbon container transportation routing on the Mombasa-Bujumbura corridor. Achieves 68% lower emissions than NSGA-II at 10% cost premium. Identifies $55/ton as optimal carbon price with 48.8% decarbonization potential. # Multi-Objective Optimization of Low-Carbon Container Routing Using PA-NSGA-II **Python** • **MIT License** • **Paper** • **Data** A Parameter-Adaptive NSGA-II Framework for Low-Carbon Container Routing on the Mombasa–Bujumbura Corridor. --- ## Overview This repository contains the complete code, data, and analysis for the paper: > **"A Parameter-Adaptive Multi-Objective Evolutionary Algorithm for Low-Carbon Container Routing under Uncertainty"** > > *Yves Ndikuriyo, Yinggui Zhang, Dung Davou Fom* > > *School of Traffic and Transportation Engineering, Central South University, Changsha, China* > > *Computational Optimization and Applications (Under Review)* ### What is LC-CTRP? The **Low-Carbon Container Transportation Routing Problem (LC-CTRP)** optimizes multimodal freight routes, balancing three conflicting objectives: | Objective | Description | Unit | |-----------|-------------|------| | **Cost** | Total transportation + transshipment + carbon costs | USD | | **Emissions** | CO₂ emissions from road, rail, and waterway transport | kg CO₂ | | **Time** | Total transit time including border delays and transshipment | hours | ### What is PA-NSGA-II? The **Parameter-Adaptive NSGA-II (PA-NSGA-II)** is a hybrid multi-objective optimization framework that integrates: ``` ┌─────────────────────────────────────────────────────────────────────────────┐ │ PA-NSGA-II Framework │ ├─────────────────────────────────────────────────────────────────────────────┤ │ ┌──────────────┐ ┌──────────────┐ ┌──────────────┐ │ │ │ Adaptive │───▶│ Size- │───▶│ NSGA-II │ │ │ │ PPO │ │ Adaptive │ │ Pareto │ │ │ │ Initialization│ │ Config. │ │ Search │ │ │ └──────────────┘ └──────────────┘ └──────────────┘ │ └────────────────────────────────────────────────────── …