Cognitive energy with Woodside
Woodside, Australia’s largest independent oil and gas company, is scaling the knowledge of engineers working in control rooms 70 miles offshore. Recognized globally for their success in oil exploration, development, production and supply, this 62-year-old organization has been breaking ground for years, and not just with their drilling rigs. The strategy behind Woodside’s success is in the way they grow and empower their workforce. The offshore energy industry itself is demanding, requiring real-time, 24/7 monitoring from crew members. With some of the largest structures on Earth, built in some of the most remote parts of the ocean, Woodside’s employees need real-time access to critical information to keep things running smoothly and ensure the safety and operational efficiency of all their employees and oil rigs.
Photo: Woodside offshore platform
Solve the brain drain
Brain drain is a challenge across the energy and utilities organizations. Woodside is no different: the company started losing swaths of irreplaceable corporate memory as older engineers retired, taking their instincts and experience with them. Although Woodside has been archiving their employees’ reports, decision logs, and technical evaluations for decades, the organization needed to replace its skilled workforce, and get the new recruits up to speed in a short amount of time.
Make everyone an expert
Woodside’s objective to make every employee within the company an expert – from the newest recruit to the most seasoned field staff, meant they needed to take 30 years of knowledge and create a living repository which could be accessed by anyone in the company, from anywhere. But having every new employee comb through technical evaluations, employee reports and decision logs would have been time-consuming, costly, frustrating and in the end also ineffective.
Enlist the help of Watson
Woodside turned to Watson and IBM to help. Using Watson technologies and APIs to search and analyze historical, scientific and experiential materials quickly and accurately, surfacing answers and critical information whenever employees need it.
Not only does Watson serve as the much-needed resource to newly hired employees, but it also provides Woodside employees with additional insights hidden within the massive archives of their institutional information to enable better decision-making.
Put tribal knowledge into energy employee’s hands
Working with IBM, Woodside developers identified the four Watson APIs needed to craft an architecture and build an intuitive design that lets engineers find the advice they need. Watson ingested the equivalent of 600,000 pages of information, or 30 years of knowledge in less than 3 seconds.
To do this, Woodside Energy combined three different Watson APIs: Natural Language Classifier, Retrieve and Rank, Watson Conversation Service.
Natural Language Classifier: This API allows users to search a corpus by asking questions as if they were talking to another person. Watson uses Natural Language Classifier to parse out the intent of a question even if it is asked in different ways.
Retrieve and Rank: After understanding the question, Watson retrieves all relevant information from the corpus, ranks them in terms of relevance, and responds with the best matches, as well as related points of inspiration.
Conversation: By incorporating a human tone, Conversation creates a better user experience and allows Watson to interact with engineers in their own language.
Create a living repository: lessons learned
Woodside named the new repository Lessons Learned. The repository enables any employee the opportunity to access Watson. The way the employees interact with Watson is through an engagement advisor called Lessons Learned – from any platform – mobile, laptop, even using robotics and speaking into a screen.
The way the employees interact with Watson is through an engagement advisor called Lessons Learned. Lessons Learned essentially enables employees to access Watson from any platform – mobile, laptop, even through robotics and speaking into a screen.
Employee use the Lesson’s Learned engagement advisor, ask Watson a question, Watson combs through the repository, ranks the data, and then provides the best match answer – all in natural language. What makes this different from something like Google search is that the Watson system is contextual – it understands intent – and is not just offering key word search.
This is important for Woodside staff because while they are working, as they detect an issue – hear something that doesn’t sound right in one of the pumps for instance, they can query Watson in a contextual environment and get the answer which matches their conditions. Here’s an example of how that works in practice:
A project engineer who is in training is facing a problem with birds on the helicopter landing pad. The engineer asks Watson, ‘What design features do we have in place on the offshore platform to deter these birds?’ The system runs the query and comes back with a case study and a solution – one which had been created by a project team ten years ago. Without Watson, it would be difficult to find the paper work from the project team that worked on the same issue ten years ago – quickly, or possible ever.
Reduce time spent searching for expert knowledge by 75%
In addition to reducing the amount of time required to locate the expert advice by 75%, the organization also made massive gains in efficiency through regular and easy access to the best practices from the past 30 years. In total, the use of Watson at Woodside helped save $10M USD.
Watch the Woodside Cognitive Energy video to discover the internal challenges that pushed Woodside to partner with IBM, and how cognitive thinking is helping them reach their goals.
Woodside Energy and IBM Watson