Green Distribution Route Optimization of Medical Relief Supplies Based on Improved NSGA-II Algorithm under Dual-Uncertainty

Shuyue Peng, Qinming Liu, Jiarui HuSustainability 2023 · DOI

TL;DR

This paper tackles the critical challenge of optimizing medical relief supply distribution during emergencies by developing a novel multi-objective model that balances transportation costs and risks under uncertain conditions. They introduce an improved genetic algorithm (MHNSGA-II) that significantly outperforms the standard NSGA-II, demonstrating substantial reductions in both cost and risk.

What Problem Does This Paper Address?

In the face of public health emergencies like pandemics, efficient and timely distribution of medical supplies is paramount. Traditional logistics models often fall short due to inherent uncertainties in demand and transportation conditions (e.g., traffic congestion, accidents). This paper addresses the need for a robust optimization framework that can:

  • Minimize costs: Including not just fixed and fuel costs, but also carbon emissions and penalties for late deliveries.
  • Minimize risks: Specifically, accident risk and congestion risk, which can delay critical aid.
  • Handle uncertainty: By incorporating fuzzy logic for uncertain supply demand and traffic flow.
  • Optimize routes and facility locations: To ensure efficient delivery from rescue centers to demand points.

The existing literature, while addressing aspects of emergency logistics and sustainable transportation, has not fully integrated these dual objectives (cost and risk) under dual uncertainty (demand and traffic) with an advanced optimization algorithm.

Methods

The core of this research lies in a two-pronged approach: a novel optimization model and an improved algorithm to solve it.

1. Multi-Objective Distribution Route Optimization Model:

  • Dual Uncertainty: The model explicitly accounts for uncertainty in:
  • Supply Demand: Represented using triangular fuzzy numbers.
  • Traffic Flow: Also modeled using triangular fuzzy numbers.
  • Dual Objectives: The optimization aims to simultaneously minimize:
  • Transportation Cost (C): This includes vehicle fixed costs, fuel costs, carbon emission costs (using a carbon tax), and time costs (penalties for delays).
  • Transportation Risk (E): This is a composite risk considering accident probability, traffic flow, accident area, and congestion-induced delays.
  • Fuzzy Constraints: Fuzzy chance constraints are used to represent the uncertain demand and traffic flow, which are then defuzzified into deterministic constraints using fuzzy credibility theory.

2. Improved NSGA-II Algorithm (MHNSGA-II):

The authors propose a "multi-strategy hybrid improved nondominated sorting genetic algorithm" (MHNSGA-II) to tackle the complexity of the multi-objective problem. Key improvements over the standard NSGA-II include:

  • Two-Stage Chromosome Real Number Coding: A novel encoding scheme for representing solutions.
  • Improved Population Initialization: An "individual insertion method" is used to generate more effective initial populations.
  • Enhanced Crowding Distance Selection: Instead of solely relying on crowding distance for selection within a Pareto rank, they use a "roulette wheel" method that gives individuals with greater crowding a higher probability of selection, aiming to maintain diversity.
  • Adaptive Crossover and Mutation Probability: These probabilities are dynamically adjusted based on population fitness, preventing premature convergence.

Datasets

The study utilizes a case study based on the real situation in Shanghai, China.

  • Candidate Rescue Centers: Locations were determined using straight-line distance and the entropy weight method (EWM). Ten candidate centers were identified.
  • Demand Points: 55 demand points were selected, corresponding to administrative districts and street-level data.
  • Medical Supplies: The paper specifies that supplies include essential medicines, disinfectants, and improvised medical equipment.
  • Fuzzy Demand Data: Fuzzy demand values (e.g., (104, 137, 170)) were generated for each demand point, reflecting uncertainty.
  • Traffic Flow Data: Fuzzy traffic flow parameters were also defined.

No new benchmark datasets were introduced; the study uses real-world geographical data for Shanghai.

Key Insights and Ideas

  • The "Trade-Off" is Real: The Pareto frontier clearly shows that minimizing transportation cost and minimizing transportation risk are conflicting objectives. Decision-makers must choose a solution that best fits the urgency and context of the emergency.
  • Fuzzy Logic is Crucial for Real-World Uncertainty: Using fuzzy numbers to model demand and traffic flow is essential for creating realistic and actionable optimization models in dynamic emergency scenarios.
  • Algorithm Enhancements Matter: The improvements to NSGA-II (MHNSGA-II) are not just incremental; they lead to demonstrably better solutions, suggesting that algorithmic innovation is key to solving complex multi-objective problems in logistics.
  • Vehicle Usage is a Sensitive Parameter: The number of vehicles used significantly impacts both cost and risk, highlighting the need for careful fleet management and allocation.
  • Credibility Matters: Increasing the credibility (or confidence) in the fuzzy constraints leads to higher costs and risks, indicating that higher certainty comes at a price.

Key Results

  • Significant Improvement with MHNSGA-II: Compared to the original NSGA-II, the proposed MHNSGA-II algorithm achieved:
  • 42.55% decrease in transportation cost.
    • 5.73% decrease in transportation risk.
  • Pareto Optimal Solutions: The MHNSGA-II generated a larger and more uniformly distributed set of Pareto optimal solutions, offering decision-makers a richer set of choices.
  • Sensitivity Analysis:
    • Vehicle Usage: Fewer vehicles generally lead to higher costs but lower risk.
  • Risk Impact Factors: Adjusting the weights of accident risk vs. congestion risk significantly impacts the trade-off between cost and risk.
  • Credibility: Higher credibility in fuzzy constraints leads to increased costs and risks.

Limitations and Future Work

  • Subjective Factors Ignored: The model primarily focuses on objective factors like time and cost, neglecting subjective human psychological factors (e.g., urgency perception of demanders).
  • Data Accessibility: Real-time data for emergency situations can be difficult to obtain, leading to the use of fuzzy numbers as approximations, which may deviate from the actual scenario.
  • Material Scheduling: The paper acknowledges that the number of required supplies at each demand point can vary, and further research into material scheduling alongside route optimization would be beneficial.
  • Demand Urgency: Incorporating the urgency of medical supplies and the psychological factors of demanders into route selection and optimization is suggested for future work.

Summary updated

Abstract

With growing concerns about environmental issues, sustainable transport schemes are receiving more attention than ever before. Reducing pollutant emissions during vehicle driving is an essential way of achieving sustainable transport plans. To achieve sustainable transport and reduce carbon emissions, on the premise of ensuring rescue timeliness, this research proposes a multi-objective distribution route optimization model considering the minimization of transportation cost and transportation risk under dual-uncertainty constraints, providing a practical framework for determining the optimal location of rescue centers and distribution routes in emergencies using fuzzy theory. First, this paper proposes objective functions that innovatively take into account the congestion risk and accident risk during the distribution of medical supplies while introducing the carbon emission cost into the transportation cost and using the fuzzy demand for supplies and the fuzzy traffic flow on the roads as uncertainty constraints. Then, this paper designs a multi-strategy hybrid nondominated sorting genetic algorithm (MHNSGA-II) based on the original form to solve the model. MHNSGA-II adapts a two-stage real number coding method for chromosomes and optimizes the population initialization, crowding distances selection, and crossover and mutation probability calculation methods. The relevant case analysis demonstrates that, compared with the original NSGA-II, MHNSGA-II can decrease the transportation cost and transportation risk by 42.55% and 5.73%, respectively. The sensitivity analysis verifies the validity and rationality of the proposed model. The proposed framework can assist decision makers in emergency logistics rescue.
BibTeX
@inproceedings{peng2023green,
  author = {Shuyue Peng and Qinming Liu and Jiarui Hu},
  title = {{Green Distribution Route Optimization of Medical Relief Supplies Based on Improved NSGA-II Algorithm under Dual-Uncertainty}},
  booktitle = {Sustainability},
  year = {2023},
  doi = {10.3390/su151511939},
}
Figure