Dan Jeavons

General Manager, Data Science, Shell

Supernova Award Category: 

Data to Decisions

The Organization: 

Shell has been an early adopter of artificial intelligence (AI), as it attempts to speed up its digital transformation and energy future. From machine learning to computer vision, deep learning to virtual assistants and autonomous vehicles to robotics, Shell has been focused on a range of technologies that have supported advances in AI. Shell has developed an integrated digital program covering core value areas – subsurface and wells, asset management and customer centricity. The program is governed by the executive committee, with senior executives taking an active role in shaping the program. Each of these themes have identified high priority digital initiatives to focus on, many of which rely on AI.  

The Problem: 

Shell is an integrated energy company going through a significant transition as the world’s need for more and cleaner energy solutions requires significant changes in the way in which energy provided. Digital technology, including AI, is a significant enabler for this transformation, from more efficient exploration and production, more reliable manufacturing, more nimble trading, and a more intimate customer experience and new digital solutions for emerging areas such as power and hydrogen. Embedding AI in every part of our organization is crucial—from making our existing businesses more effective, efficient and make us competitive.

Anyone can develop a small-scale proof of concept with a machine learning model that meets a specific local requirement. But, to make an impact and maintain it at scale, we need to develop solutions which can be deployed globally at a rapid pace.

The Solution: 

To enable AI at scale across Shell’s businesses, we are standardizing approaches and aligning on common data structures, platforms, tools and ways of working. This allows us to share best practice, code and work seamlessly in cross discipline global teams.

Shell.ai is a change program formed to drive a common approach to data science platform technology, develop consistent ways of working, and build a community of practice which can demonstrate the art of the possible, and share best practices across the entire business.

The Shell.ai platform provides a framework for new digital artificial intelligence solutions and capability developments and is a focus for collaborative projects. Shell has worked with technology partners such as C3.ai and Microsoft to develop solutions that scale and can be replicated globally.

The Results: 

Our team of data scientists is delivering AI solutions which are realizing significant value across the energy value chain. Early successes enabled by AI include technologies such as well location optimization, predictive maintenance, real-time production optimization and customer loyalty schemes like Go+.

For example, in one location alone, in Europe’s largest oil refinery, predictive maintenance has improved safety and saved millions of dollars through proactive interventions. In the same location, we have deployed both robots and drones to accelerate the manual inspection routines and identify changes in equipment. Supported by Shell.ai, engineering leaders at this location have led the charge to embed AI in their daily operations. Data Scientists working alongside Engineers are working to resolve issues specific to the facility. Additionally, they are helping the broader asset community understand the impact that AI can have.

Furthermore, leaders from the asset are actively evangelizing the solutions to other parts of Shell. In the example of predictive maintenance, we were already monitoring thousands of pieces of equipment across 23 operating assets in our Upstream, Downstream Manufacturing and Integrated Gas assets.

Finally, a community of practice has led to over 3000 professionals who share knowledge across the organization. This is having a significant impact in terms of business pull for more AI solutions.

Metrics: 
  • Digitalization has already enabling more than 1 billion Euros of cost reductions, production increases and increased customer margins across Shell’s businesses in 2019 alone and AI has been a critical enabler.  
  • We have a community (Shell.ai community) of over 3000 people engaged the Shell.ai program. We have over 160 AI projects in flight. Our Shell.ai Self-Service platform is used by over 800 citizen data scientists.
  • We have held over 40 hackathons in the last few years.
  • We have over 1.1 trillion rows of curated sensor data in our data lake – used for the purposes of machine learning. Over the last year, we have quintupled the amount of data we have integrated into our central data platforms – handling over 850,000 sensors producing data at a frequency of once every minute or higher. We now handle 400 GB of data every month and cover more than 70 PB of storage. 
  • We are managing in excess of 20,000 live machine learning models, monitoring over 1500 different pieces of equipment
The Technology: 

The Shell.ai platform is an important enabler for Shell’s transformation. We leverage a citizen development toolset based on a number of open source technologies based on the Microsoft Azure cloud. We are using the C3 AI Suite for scaling and replicating large scale solutions. Underpinning is a massive scale data lake which ingests trillions of rows of time series data into Azure Data lake store.

Disruptive Factor: 

Shell formed a data science centre of excellence to centralize a coordinated effort to deploy AI in all aspects of the energy business. Dan Jeavons was appointed General Manager for data science in 2018, leading the Data Science CoE and reporting into the technology organization to ensure successful adoption of the technology as part of a business change program. Dan has been a catalyst internally and an evangelist externally to promote Shell’s progress in this digital transformation.

Key innovations:

  • Select foundational technologies to shift from descriptive and diagnostic analytics to predictive, prescriptive, and AI-driven insights
  • Showcase what is possible with those technologies, especially evolving from traditional business intelligence and analytics to real-time advanced analytics
  • Facilitate the sharing of best practices across Shell

Shell also separated innovation from operational support by creating a separate business service center to handle ongoing operational management and maintenance of advanced analytics deployments so the CoE could focus on innovation and alignment.

Shining Moment: 

We moved from a selection of proof of concept and pilot projects to achieving true production scale for AI. Since then, we scaled our efforts while building a team and delivering AI projects across the business. Importantly we captured the imagination by showcasing the art of the possible and driving a change across our business. We are accelerating the change to drive even further value from AI across Shell.

About Shell

Royal Dutch Shell plc companies have operations in more than 70 countries and territories with businesses including oil and gas exploration and production; production and marketing of liqueed natural gas and gas to liquids; manufacturing, marketing and shipping of oil products and chemicals and renewable energy projects. For more information, visit www.shell.com.