Shell drills for data insights

Video: GPUs are game changer for oil and gas sector

Energy and petrochemical company Shell is no stranger to using advanced technologies to enhance business processes. One of the latest efforts, by the corporation's Shell UK company, is to leverage advanced data analytics to monitor operations and improve its supply chain decisions and inventory.

"We are working towards the creation of integrated teams whereby our scientists, engineers, business experts, and IT specialists work in close collaboration to deliver our projects," said Dan Jeavons, general manager, Advanced Analytics Centre of Excellence at Shell.

The move to advanced analytics began several years ago, and the company's top IT executive saw a clear need to create an Advanced Analytics Centre of Excellence.

Strong leadership support "allowed the team to experiment with analytics without fear of failure, [and] instilled a culture of analytics," Jeavons said. "We developed a learning culture emphasizing the importance of learning from prior projects and building on successes as well as failures."

In 2014 Shell UK began using an analytics platform from Alteryx, with initial use cases around supporting the deployment of open source algorithms and for cross-selling opportunities among customers.

The company has launched several analytics efforts that have made a significant impact on the business.

In one project, Shell's upstream operations team uses predictive analytics capabilities to optimize the ordering, storage, and utilization of pieces of spare part inventory for onshore and offshore oil rigs. These include well heads and pipeline parts. The project has delivered millions of dollars in benefits and paid for itself in less than four weeks, Jeavons said. Shell plans to deploy the technology across its downstream business as well.

In another, Shell's downstream lubricants supply chain operations used analytics to develop a suite of tools that provide critical information on inventory, margin, forecast accuracy, and blending options. "These tools have taken manual processes built in [Microsoft] Excel and R into fully automated, end-to-end workflows," Jeavons said. That has made quality data available to users much more quickly than was previously possible.

In addition, Shell's downstream trading compliance team uses analytics to help monitor operations across multiple markets, achieving compliance with current regulations and creating transferability as markets and regulations develop.

One of the biggest challenges with data analytics has been adoption -- how to ensure that the products it produces are usable and relevant to the people across the business who use them every day. "Our biggest learnings are around engagement and ownership," Jeavons said. "If they feel they own it and they built it, they are far more likely to use it. We try to embed the users in our development processes."

Another issue as the analytics team grows is managing the datasets it's developing. "As more use cases emerge, we are increasingly developing a data asset which we know is already generating business value," Jeavons said. One frustration "has been our lack of ability to generate reuse."

The team has been part of the beta program for Alteryx Connect and has already seen the benefits of data engineers sharing their data sets with each other. "We've already seen removal of duplication and reuse of data sets that had already been developed," Jeavons said. "The potential to create a value-focused approach to developing a common data asset is a big opportunity -- but remains a big challenge."

PREVIOUS AND RELATED COVERAGE

As oil and gas embrace automation, five things to consider

Robotic process automation can enhance processes, but oil and gas businesses need to consider certain factors before deploying the technology.

Will energy companies strike oil with AI?

Adoption of artificial intelligence is still sluggish, but business can reap big benefits.

GPUs: Key to success for oil and gas companies

More powerful computing enables businesses to gain value from the huge amounts of data they're collecting.