Woodside Energy in Western Australia has been using artificial intelligence since 2015. The company has spent a decade building predictive models and optimisation systems for exploration, drilling, and plant maintenance. It is now moving toward agentic AI to support complex industrial workflows.
Operational data as the base layer
Woodside’s approach began with traditional analytics rather than consumer-facing chatbots. Vice president for digital Andrew Melouney notes the company has always handled large volumes of data from equipment and assets.
“Those have created really clear, quite high-value use cases for us.”
The focus remains on reliability, safety, and efficiency. Melouney says the firm has applied machine learning techniques to its data sets since around 2015. Recent advances in generative AI now sit on top of that existing foundation.
Augmenting operators, not replacing them
The company designs systems to support expertise in high-stakes environments rather than replacing human operators. A prime example is the “Startup Advisor,” an AI copilot that helps manage the complex process of starting liquefied natural gas (LNG) plants.
“We’re really thinking about, how does it support the people in the organization in terms of empowering them to make better decisions, to make faster decisions,” Melouney explains.
Work involves ensuring LNG plants start up reliably, consistently, and safely. The aim is to give frontline workers the tools required to do their jobs.
Standardised platforms and governance
Woodside is graduating from isolated experiments to enterprise-wide systems built on standardised platforms and governed data. The transition requires rethinking technology stacks and how work gets done.
“We’re not just bolting AI onto an existing process,” he says. “We’re deeply thinking about how that work needs to be reimagined.”
Melouney’s motto is “Think big, prototype small, and scale fast.” The company spends time teaching people how to work in agile ways, use design thinking, and solve problems effectively.
Maintenance intelligence results
Data acts as an asset for the company. Woodside operates facilities with sensors streaming real-time data. The firm invested years in an enterprise-scale data platform to ensure security and trust.
“Maintenance intelligence” is one solution. It analyses historical maintenance records alongside equipment performance. By keeping data well-governed and in one place, the system correlates maintenance records from SAP with equipment performance from a time series data lake.
The tool recommends the optimal timing for maintenance activities. The goal is simple: do the right work at the right time. Pilots on one asset have shown a reduction in maintenance hours of up to 15% over five years.
Agentic AI now sits on top of those analytical models to provide better insights. The company ensures asset and operational teams retain accountability for final decisions.
What it means
Success depends on the operational foundations built beneath the hype. Companies that spent years collecting data, governing it, and training staff are positioned to scale AI effectively. The ambition is an autonomous enterprise where agents interact deeply with core workflows while humans remain in the loop.
