Many organisations are exploring how AI, analytics, and digital technologies can support better operational decisions, but turning potential into practical results is rarely a straightforward process. At AGL, we started with a simple question: where can data help solve real operational problems? The journey since has involved experimentation, learning, setbacks, and a growing understanding of what it takes to deliver value at scale.

In this session, we share several practical use cases that have emerged from that journey, including detecting boiler tube leaks earlier, reconstructing missing production data following sensor failures, and applying analytics to better understand and manage geotechnical risk events such as tunnel collapse. We will also discuss emerging optimisation opportunities within hydro operations.

Rather than presenting a polished end-state solution, this session provides a candid account of how we got started, what has worked in practice, and where challenges remain. We'll talk through the journey—warts and all—including lessons learned, organisational barriers encountered, and the approaches that are helping build confidence, capability, and momentum for broader adoption across the business.

1. How to identify and prioritise practical analytics and AI use cases that deliver operational value and build organisational support.
2. Common challenges encountered when moving from proof-of-concept to real-world implementation, and the lessons learned along the way.
3. Strategies for building capability, trust, and momentum to scale data-driven decision-making across operations and engineering teams