DAVID COTTINGHAM AI IMPLEMENTATION
THE IMPORTANCE OF HIGH-QUALITY NETWORK DATA AI IMPLEMENTATION: Many operators are looking to capitalise from AI because of the promise of enhanced efficiency and customer experiences. However, to enjoy these benefits, the successful implementation of AI depends on the quality of underlying fibre network data. Operators need to prepare their data before AI implementation to reduce inaccuracies that could result in network failure and severe disruptions, writes David Cottingham , CTO at IQGeo.
A ccording to market ahead of competition, it’s clear that telecom operators need to integrate emerging technologies, putting growing pressure on operators to implement AI. A report by McKinsey titled “The AI- native telco: Radical transformation to thrive in turbulent times” highlights the significant impact Artificial Intelligence research firm Valuates, the global AI in telecom market is expected to reach almost US$15 billion by the end of 2027. To stay (AI) and Machine Learning (ML) can have on growth and disruption in the telecommunications sector. THE IMPORTANCE OF NETWORK DATA QUALITY At the core of any telecom business is its physical network infrastructure. Operational efficiency and future technological developments rely on accurate data about network asset locations, physical connections (such as fibre and splice boxes) and network configurations. Inaccurate or incomplete network data results can lead to operational inefficiencies which will increase costs and impact overall decision-making. This will significantly hinder the ability of telecom operators to compete in a fast-paced market. Inefficient operations lead to wasted resources and slower
service delivery, ultimately reducing that all-important ROI. Poor decision-making, driven by inaccurate or incomplete data, can have a large impact to business results overall, even affecting investment decisions. Together, these challenges undermine operational effectiveness and put a strain on financial resources. To actualise the promised benefits of AI and ML implementation, telecom operators need a reliable foundation of high-quality network asset data. Otherwise, as the adage goes, “garbage in, garbage out”. To avoid this, operators should prioritise strategies and technologies aimed at creating a complete and accurate network model. This includes: 1. Comprehensive Asset Management: Effective asset management involves accurately tracking the geospatial location and connectivity architecture of all assets within a network. This ensures that each asset, whether physical or virtual, is properly mapped and monitored. 2. Maintaining Network Documentation: Network documentation requires continuous updates to reflect changes in infrastructure. As troubleshooting evolves, it’s crucial to invest in documentation tools and practices. This will help handle the regular
upgrades and expansions within the industry, reducing inefficiencies and ensuring smooth network management. This means that system changes are tracked in real-time. 3. Integrated Systems: Transitioning from heterogeneous systems to a unified solution will help manage the fibre network lifecycle. Integrated systems consolidate data, from planning and deployment to maintenance and monitoring, creating a single source of truth. Integrated systems therefore lead to better decision-making and more efficient network management.
MAKING NETWORK OPTIMISATION A PRIORITY At the core of every telecom
organisation lies its physical network. Therefore, it’s essential that operators effectively manage their networks to accelerate ROI. One of the first steps in this process is to replace spreadsheets and CAD drawings with fibre network documentation software. Many larger established broadband operators are still using outdated GIS systems and a collection of siloed applications. This suggests that more agile operators, who have implemented integrated lifecycle management solutions tailored for fibre networks, could outperform large broadband providers.
16
www.opticalconnectionsnews.com
| ISSUE 40 | Q1 2025
Made with FlippingBook Online newsletter maker