Three years ago, Woodside Petroleum kicked off an initiative to determine if it could use analytics techniques to predict the failure of a particular piece of plant equipment.
As a cheap and quick test was promising, Woodside's foray into analytics "exploded" from there, setting up an internal team comprising engineers, asset specialists, and others to focus on the opportunities data science presented. It was important to the company to have the team built from business leaders, rather than technical specialists.
Woodside's chief digital officer Tom Ridsdill-Smith told the Gartner Symposium/ITxpo on the Gold Coast on Monday that this team straight away turned to open source tools, and partners that were excelling at data-driven platforms, as Woodside was aware of its capabilities.
Working with IBM Watson, Woodside built a customised tool that allowed its employees to find detailed answers to highly specific questions -- even on remote oil and gas facilities. Watson ingested the equivalent of 38,000 Woodside documents and the initiative soon morphed into Willow.
The idea of Willow was to create a single chat-like interface that worked across all of Woodside's systems.
Powered by Watson on IBM Cloud, Ridsdill-Smith explained that Woodside's employees can ask Willow questions in natural language, such as: "What is the maximum weight of a helicopter landing on the platform?", and Willow will respond accordingly.
"The quickest and most effective way that we found to build data science tools is taking a very well-defined business problem and we build a tool around that," he explained. "Willow is a prototype for us, and we've experimented with different forms of it."
Willow, in its current form, has about 500 users since it was launched only a couple of weeks ago, even though it was only released to about 200.
"It's escaped captivity -- it's gone out into the wild," he said. "For us, that's a success, it means that it's valuable and people want to be able to use the tool."
Woodside built Willow in a modular fashion, but Ridsdill-Smith said there are a couple of dozen more parts that Woodside will build into it once the organisation is satisfied that its performance is going to scale.
"Willow really is a tool built around our vision to build the best cognitive intelligence for the organisation," he added. "We don't do technology for technology sake."
"Ultimately, what it's done is help drive an innovative culture within Woodside, but most importantly it's helped us solve some business problems."
Woodside's vision around data science is that by 2018, it will deliver a collective intelligence solution to the entire organisation.
Disclosure: Asha Barbaschow travelled to Gartner Symposium/ITxpo as a guest of Gartner.
PREVIOUS AND RELATED COVERAGE
Drowning in data is not the same as big data. Here's the true definition of big data and a powerful example of how it's being used to power digital transformation.
Companies will be building the foundation for artificial intelligence, analytics, IoT, digital twins, and automated systems in 2018.
IBM's plan is to create a data science operating system that can bring together data scientists, analysts, and business leaders.