利用數位孿生技術突破自主網路中的跨域障礙

本報告檢驗了目前通訊網路自動化的局限性，尤其是在基礎設施、服務和品質要求日益複雜的情況下。在虛擬化技術快速發展和跨域互動日益頻繁的環境下，僅靠局部最佳化已遠遠不夠。域級最佳化決策實際上可能會降低整體效能，並阻礙真正端到端自主性的實現。

為了解決這種碎片化問題，本報告提案引入數位網路孿生技術，該技術能夠跨所有技術領域和服務層表示和類比網路。透過數據整合和集中化，以及清晰的依賴關係映射，數位雙胞胎為人工智慧提供了一個參考環境。這使得人工智慧能夠描述網路狀態、預測干預措施的影響，並提案多領域最佳化方案。

該分析還著重指出了相關的營運和經濟挑戰。模擬和平衡效能、容量、能耗和服務保障之間的權衡取捨，正成為降低網路變更相關風險和提升營運穩定性的關鍵手段。

目錄

第1章摘要整理

第2章 市場環境及促進因素

• 自動化的限制以及向基於意圖的管理的轉變
• 需要進行風險管理、能源最佳化和網路孿生。

第3章：技術與架構概述

• 什麼是網路數位雙胞胎，其架構原則是什麼？
• 參考架構
• 跨域網路孿生機制

第4章：價值鏈

第5章 市場規模及預測

• 通訊 AIOps 和平台
• 可觀測性與網路保障
• 通訊網路自動化
• 網路數位雙胞胎

第6章：使用案例和部署場景

• Telenet：管理計畫變更對服務和客戶的影響
• BT、AWS、Celfocus：整合保障與多領域根本原因分析
• AT&T 愛立信：規劃實施前的模擬

第7章：挑戰與機遇

• 任務
• 機會

第8章 戰略建議

• 關於策略管理的建議
• 關於網路和IT管理的建議
• 向監管機構和投資者提出的建議

The report examines the limitations of current telecom network automation, particularly amid growing complexity in infrastructure, services, and quality requirements. In an environment defined by increased virtualisation and the proliferation of cross-domain interactions, local optimisation is no longer adequate: decisions optimised at the domain level may actually undermine overall performance and impede the development of true end-to-end autonomy.

To address this fragmentation, the report proposes introducing a digital network twin capable of representing and simulating the network across all technical domains and the service layer. Through the federation and unification of data, along with an explicit mapping of dependencies, this digital twin provides artificial intelligence with a reference environment. This enables AI to explain network states, predict the impact of interventions, and recommend multi-domain optimisations.

The analysis also underscores the operational and economic challenges involved. The capability to simulate and balance trade-offs between performance, capacity, energy consumption, and service commitments is becoming a critical lever for mitigating risks related to network changes and for enhancing operational stability.

Table of Contents

1. Executive Summary

2. Market context and drivers

• 2.1. The limits of automation and the shift towards intent-based management
• 2.2. Risk management, energy optimisation and the need for a network twin

3. Technology and architecture overview

• 3.1. What a network digital twin is & architecture principles
• 3.2. Reference architecture
• 3.3. How the cross-domain network twin works

4. Value chain

5. Market size and forecasts

• 5.1. Telecom AIOps and platform
• 5.2. Observability and network assurance
• 5.3. Telecom network automation
• 5.4. Network digital twins

6. Use cases and deployment scenarios

• 6.1. Telenet: managing the impact of planned changes on services and customers
• 6.2. BT, AWS and Celfocus: unified assurance and multi-domain root cause analysis
• 6.3. AT&T and Ericsson: pre-deployment simulation for planning

7. Challenges and opportunities

• 7.1. Challenges
• 7.2. Opportunities

8. Strategic recommendations

• 8.1. Recommendations for strategic management
• 8.2. Recommendations for network and IT management
• 8.3. Recommendations for regulators and investors

List of Tables

4. Value chain

Table 1: Value chain for network digital twins

5. Market size and forecasts

• Figure 1: AIOps market from 2024 to 2030 in billions of euros
• Figure 2: Network observability and assurance, APM and telecoms market from 2024 to 2030 in billions of euros
• Figure 3: Global and telecoms network automation market from 2024 to 2030 in billions of euros
• Figure 4: Global and telecoms digital twins market from 2024 to 2030 in billions of euros