Scientific Reports Article in Press
Optimizing Poultry Vaccine Distribution with XAI-Enabled Intertwined Logistics
This research introduces an XAI-enabled framework for optimizing intertwined logistics in poultry vaccine distribution. It proposes a novel mixed-integer linear mathematical model for vehicle assignment and routing across multi-logistic, multi-depot, multi-vehicle networks, demonstrating significant efficiency gains and resilience under stress.
Executive Impact Summary
Our analysis demonstrates a highly efficient and resilient intertwined logistics network for poultry vaccine distribution. By strategically assigning vehicles, including planned layoffs, the system achieves an optimal cost structure with 80% shipping, 16% traveling, and 4% opportunity costs. Advanced XAI techniques reveal that shipping costs and demand are the primary drivers of total cost variability, providing actionable insights for managers.
0%
Efficiency Gain
via Layoff Strategy (Table 8)
via Layoff Strategy (Table 8)
0%
Cost Optimization
Shipping Cost Share (Figure 7)
Shipping Cost Share (Figure 7)
0%
Cost Increase Offset
for Resilient Operations (Table 9)
for Resilient Operations (Table 9)
Deep Analysis & Enterprise Applications
Select a topic to dive deeper, then explore the specific findings from the research, rebuilt as interactive, enterprise-focused modules.
Optimal Cost Structure Achieved
The proposed model successfully optimizes the total logistics costs, demonstrating high efficiency in poultry vaccine distribution. Shipping costs represent the largest component, highlighting its importance for management focus.
80%
of Total Cost is Shipping Cost
Source: Figure 7, Table 4
Source: Figure 7, Table 4
The Economic Advantage of Vehicle Layoff Strategy
Implementing a vehicle layoff strategy significantly reduces overall costs by preventing inefficient allocation of surplus vehicles. Neglecting this strategy would lead to a substantial increase in opportunity costs.
| Scenario | Total Cost (Z) | Opportunity Cost (OC) | Change in OC |
|---|---|---|---|
| With Layoff Strategy | 36,125 | 1,300 | Base |
| Without Layoff Strategy | 38,375 | 3,300 | ▲ 153.85% |
Network Resilience Under Fleet Disruption
The intertwined logistics network exhibits strong resilience. When two vehicles became inaccessible, the system successfully adapted by implementing a vehicle-sharing strategy between logistics, ensuring continued vaccine distribution. This adaptability, however, comes with an increased traveling cost.
Disruption Scenario: Two Vehicles Lost
Scenario: Inaccessibility of two vehicles from Logistic 1, Depot 2.
Solution: A vehicle from Logistic 2, Depot 1 was dynamically assigned to Logistic 1, Depot 2 to fulfill demand.
Impact: Total costs increased by 30%, primarily due to a 186% increase in traveling costs, but the network remained operational.
Key Takeaway: Intertwined logistics enhance resiliency but require careful management of increased operational costs during disruptions. Source: Table 9, Section 6.3
XAI-Based Global Sensitivity Analysis Framework
Our framework leverages explainable AI (XAI) to perform global sensitivity analysis on complex mathematical models, providing interpretable insights beyond traditional linear regression approaches.
Enterprise Process Flow
Propose significant parameters and outputs
→
Propose levels of parameters and standardize values
→
Propose fractional or full factorial designs
→
Solve the model per pattern to provide dataset
→
Train interpretable ML models per output
→
Select the most accurate algorithm
→
Derive sensitivity measures (e.g., SHAP, LIME)
→
Prescriptive analysis (Managerial insight)
Shipping Cost & Demand: Key Cost Drivers
XAI-based global sensitivity analysis using LightGBM reveals that shipping costs are the most significant factor influencing total costs, followed by customer demand. This insight guides where to focus optimization efforts for maximum impact.
Shipping Cost
Most Significant Impact on Total Costs Variability, followed by Demand
Source: Table 14, Figure 10, Section 6.4.2
Source: Table 14, Figure 10, Section 6.4.2
Superior Accuracy of XAI Models for Metamodeling
Complex mathematical models, such as vehicle assignment-routing, exhibit significant nonlinearity. Traditional linear regression fails to capture this, while advanced AI models like LightGBM provide highly accurate metamodels for sensitivity analysis.
| Model Type | Overall R² Accuracy | Suitability for Nonlinearity |
|---|---|---|
| LightGBM (XAI) | 0.9984 | High |
| Linear Regression (Traditional) | 0.7993 | Low |
Advanced ROI Calculator
Estimate the potential return on investment for implementing an XAI-enabled intertwined logistics solution in your enterprise.
Estimated Annual Savings
$0
Annual Hours Reclaimed
0
Accelerate Your AI Implementation: The OwnYourAI Roadmap
Our structured approach ensures a seamless integration of advanced AI analytics into your existing logistics operations, maximizing impact with minimal disruption.
Phase 1: Discovery & Assessment
Comprehensive analysis of your current logistics infrastructure, data availability, and strategic objectives to identify key areas for AI intervention.
Phase 2: Model Design & Development
Custom design and development of XAI-enabled mathematical models tailored to your specific supply chain and distribution challenges, including vehicle assignment, routing, and sensitivity analysis.
Phase 3: Integration & Deployment
Seamless integration of the developed AI models with your existing enterprise systems, followed by rigorous testing and phased deployment to ensure stability and performance.
Phase 4: Training & Optimization
Empowering your team with the knowledge and tools to effectively utilize the new AI systems, coupled with continuous monitoring and optimization for sustained performance gains and adaptability.
Ready to Transform Your Logistics?
Leverage cutting-edge XAI to build resilient, efficient, and explainable logistics networks. Our experts are ready to guide you.
Ready to Get Started?
Book Your Free Consultation.
Let's Discuss Your AI Strategy!
Lets Discuss Your Needs
Select a Date
Sun
Mon
Tue
Wed
Thu
Fri
Sat