Aerospace & Defense Supply Chain

Aerospace & Defense Industry
Global Supply Chain Analysis

A comprehensive network analysis based on 4,620 entities and 8,134 relationships extracted from 3,000 news articles

4,620 Entities
8,134 Relationships
3,733 Largest Connected Component
2.81 Average Suppliers
01 Overview

In this project, we analyzed 3,000 news articles on the aerospace and defense industry collected via the EventRegistry API to reveal the structure of the global supply chain network. Using parallel processing with Gemini 2.5 Flash Lite, we extracted entities such as companies, government agencies, universities, and research institutes, along with the relationships among them—transactions, partnerships, technology transfers, and more. After semantic deduplication, this yielded a high-quality network dataset.

02 Key Findings

1. Scale-Free Network Structure

The analysis revealed that the aerospace and defense supply chain exhibits a scale-free network structure. A small number of hub entities (NASA: 236 connections, SpaceX: 245 connections) have exceptionally high connectivity, while the vast majority of nodes have only about 2–3 connections. This structure is formed by the preferential attachment mechanism described by the Barabási-Albert model.

2. Vulnerability from Low Redundancy

The average number of suppliers is 2.81 and the average number of supplier countries is 1.55—both extremely low—indicating a lack of supply chain redundancy. While this is the result of efficiency-focused optimization, it heightens vulnerability to geopolitical risks and natural disasters.

3. Institutional Causal Mechanisms

The analysis made clear that institutions and policies are key factors determining network structure—for example, the cascade effect of ITAR (International Traffic in Arms Regulations) regulations, the de facto standardization of U.S. standards through NATO standardization, and the structural supply chain integration driven by India's Offset policy (30% mandate).

4. Cross-Industry Heterogeneity

The commercial aircraft industry (Boeing/Airbus duopoly, $32B development costs), the space industry (SpaceX monopoly, unintended consequences of NASA procurement policy), and the defense industry (monopsony buyers, dual-use technology) each have different market structures and barriers to entry, exhibiting a complexity that no single policy can address.

03 Contents

Interactive Network Visualization →

A dynamic network visualization built with D3.js. Filterable by entity type, country, and relationship type. Node sizes are optimized on a logarithmic scale to ensure readability.

Comprehensive Policy Analysis Report →

Policy recommendations grounded in deep institutional analysis that integrates the Edler & Georghiou policy typology, Mazzucato's entrepreneurial state theory, and network theory, viewed through the lens of economic security and supply chain resilience. It presents concrete short-, medium-, and long-term roadmaps.

Methodology in Detail →

An explanation of the entire process: data collection (EventRegistry API), entity extraction (parallel processing with Gemini 2.5 Flash Lite), semantic deduplication, network analysis (NetworkX), and visualization (D3.js).

04 Technical Stack
  • Data Collection: EventRegistry API (3,000 articles, 100+ sources)
  • Entity Extraction: Google Gemini 2.5 Flash Lite (100 workers, parallel processing)
  • Deduplication: Gemini semantic similarity assessment (5,170 → 4,620 entities)
  • Network Analysis: NetworkX (centrality, community detection, inter-country flows)
  • Visualization: D3.js v7 (force-directed graph, logarithmic scaling)
  • Theory: Barabási-Albert, Edler-Georghiou, Mazzucato, industrial organization theory
05 Data Source

This analysis is based on 3,000 English-language news articles (from more than 100 media sources) about the aerospace and defense industry in 2024. The articles were collected from the EventRegistry API's dmoz/Business/Aerospace_and_Defense category and include the latest developments concerning major players such as Boeing, Airbus, SpaceX, Lockheed Martin, Raytheon, NASA, ESA, and ISRO.

The extracted entities are classified into six types—company, government_organization, subsidiary, region, university, and research_institute—with each entity's nationality, role, and relationships recorded in detail.

06 Policy Implications

The main policy recommendations derived from this analysis are as follows:

  • Introducing an Offset Policy: Domestic production and technology transfer mandates based on the Indian model (30% mandate, $36M threshold)
  • TRL-Staged Funding: TRL 1–3 (competitive research grants) → TRL 4–6 (SBIR-type venture support) → TRL 7–9 (government procurement)
  • Regulatory Sandbox: Accelerating the certification process for new technologies (eVTOL, reusable rockets)
  • Mandatory Stress Testing: Institutionalizing supply chain vulnerability assessments for defense contractors
  • Friend-Shoring Framework: Multilateral ITAR reform among Japan, the U.S., Europe, Australia, and India, and NATO Plus standardization

For details, please see the Policy Analysis Report.