Dan Jeavons

General Manager, Data Science , Shell

Supernova Award Category: 

Data to Decisions

The Company: 

Shell is a global group of energy and petrochemical companies. Its operations are divided into: Upstream, Integrated Gas and New Energies, Downstream. Its Projects & Technology organization manages the delivery of Shell’s major projects and drives our research and innovation.

In Upstream, Shell focuses on exploration for new liquids and natural gas reserves.

In Integrated Gas and New Energies, Shell focuses on liquefying natural gas (LNG) and converting gas to liquids (GTL) The New Energies business has been established to explore and invest in new low-carbon opportunities.

In Downstream, Shell focuses on turning crude oil into a range of refined products. In addition, we produce and sell petrochemicals for industrial use worldwide.

The Problem: 

The oil and gas industry has always been at the forefront of digital transformation. From control systems in refineries to sensors across a digital oil field to sensors in cars, data and analytics continue to be an ever-changing source of competitive advantage. As the energy market rapidly progresses and the world transitions to a lower hydrocarbon future, Shell must use data and increasingly advanced analytics as well as artificial intelligence to drive a coherent digital strategy.

But with so much data available, the company had a major challenge to address: a lot of people within the 90,000-person organization were trying to leverage their data in new and interesting ways, but there was not a lot of coordination. Shell needed structure to enable a a skilled workforce largely composed of engineers and scientists to develop digital solutions based on common platforms. In addition, Shell needed a programme of initiatives that would drive tangible bottom-line impact from digital technologies to make the case for further investment.

Today, Shell is working toward the creation of integrated teams whereby its data scientists, data engineers, business experts and IT specialists work in close collaboration in a co-located fashion to deliver digital solutions in around 3-4 months. Some of these projects are now sponsored directly by the Executive Committee.

The Solution: 

The company’s digital strategy was 4 years in the making. For example, the Advanced Analytics CoE was created in 2013, when Shell’s IT executive first saw a clear need to incubate an Advanced Analytics as part of the digital agenda, which has now broadened to form part of a wider Digital CoE encapsulating a broader skillset.

That said, the core of the mission hasn’t changed – the team is still focused on developing data-centric skillsets across Shell by:

  • Developing a set of platform technologies that allowed for advanced analytics across our group – which included deployment of technologies such as Alteryx and Databricks.
  • Showcasing the “art of the possible” by creating rapid prototypes (or Minimum Viable Products) to demonstrate value of data.
  • Facilitating best practice sharing across Shell by creating an Digital Lab environment for experimentation and an Analytics Network.
The Results: 

Previously, most of the data science tools were licensed by individual departments and used locally on powerful desktops. Shell now leverages a common, cloud-based data science platform and the associated digital capabilities to solve a multitude of business questions, from rightsizing inventory, to improving supply chain, and from predictive maintenance, to optimizing operations in downstream manufacturing units.

Shell’s journey to digital excellence focused on the deployment of self-service analytics both for the data science teams who look to deploy mission-critical models for real-time use and for the longer-term strategy of filling the “data science skills gap” – training up scientists and engineers to leverage new tools for themselves by offering structured hackathon events as well as dedicated projects to build capability beyond the CoE.

Strong leadership allowed the team to experiment without fear of failure, instilling an agile, analytical culture. Shell also developed a learning culture that emphasizes the importance of learning from prior projects and building on successes as well as failures. The team focused on developing digital products from the start – it wasn’t just about dashboards, but around providing supporting tools to help people to do their jobs.


The Analytics Network in Shell now has over 1600 members, and over 400 of those are actively building digital solutions using the technologies deployed in the Digitalisation Lab —supported by a team of 120 data scientists and data engineers within the CoE. The group has launched many efforts that have made a very significant impact on the business bottom line.

In one well publicised example, Shell's Upstream Operations team uses predictive analytics capabilities in Alteryx and Databricks to optimize the ordering, storage and utilization of pieces of spare part inventory for onshore/offshore oil rigs-ranging from well heads to pipeline parts. The project has delivered millions of dollars in benefits and paid for itself in under 4 weeks. The technology has more recently be rolled out to Shell’s Manufacturing and Chemicals businesses.

Many more examples can be found in the “Disruptive Factor” section.

The Technology: 

Alteryx, Databricks and other associated open source technologies such as R and Python are available as part of Shell’s Digitalisation Lab, which essentially serves as its data science workbench. Alteryx is quickly becoming the most popular tool in the toolbox being used by 88% of lab users across multiple Shell business units. The variety of applications is staggering – in the previous section there are just a few examples.

Disruptive Factor: 

Given its size, global reach and quantity of data assets, Shell’s journey to analytics excellence was no easy feat, but the results speak for themselves:

Shell Downstream Lubricants Supply Chain developed an award-winning suite of tools that provide critical information on inventory, margin, forecast accuracy and blending options. These tools have taken manual processes built in Excel & R into fully automated, end-to-end workflows making quality data available, significantly faster than was previously possible.

Shell Downstream Trading Compliance team developed a set of tools to help monitor operations across multiple markets, achieving compliance with current regulations and creating transferability as markets and regulations develop.

Shell Exploration had a highly manual process for analysing information coming back from drilling campaigns. They have constructed a "New Well Portal" where various subject matter experts can visit to analyze future extraction opportunities.

Shell Downstream Retail has developed a tool to support the optimal location of retail stores and associated offers at the site location by network planners. The tool provides predictive insights about the potential for a particular site at the click of a button.

Shining Moment: 

Most recently, shortly after the implementation of a predictive maintenance algorithm on the Shearwater asset, a compressor trip was successfully predicted. This was a break-through moment as it really brought home to the business the potential of data science and digital technology to radically change current operational processes.

About Shell

Shell companies have operations in more than 70 countries and territories with businesses including oil and gas exploration and production; production and marketing of liquefied natural gas and gas to liquids; manufacturing, marketing and shipping of oil products and chemicals and renewable energy projects.