Sunday, 12 July 2015

How Shell is using Analytics to drill deeper!


Introduction

The rate at which the field of Data Science and Analytics is booming, business in every industry has developed data-led strategies for overcoming problems and solving challenges. The oil and gas industries are no different. With the global energy industry facing an onslaught of new challenges, oil and gas companies no longer rely on techniques and technologies of the past.

Analytics can optimize the activities associated with exploration and production, including oilfield production forecasting, predictive asset maintenance, reservoir characterization and analytics for unconventional resource recovery.

One of the leading provider of Energy, Royal Dutch Shell, uses big data to improve its operations and to increase oil and gas output of their wells. Since a couple of years they have started lowering optical fiber cables within their wells. These cables are supplied with sensors that measure everything within the wells. With this data, Shell analysed how the wells are doing and how much oil or gas is still left.

Shell is credited with producing about 2300 barrels of oil every minute and transport fuel to 10 million retail customer daily. So far, it has generated 46 Petabytes of data through its seismic sensors that are used to discover new oil wells. Shell felt that they have a lot of data but very less decision making ability. With this in mind, there was a recognition in the IT function that Shell needs to adapt to predictive modelling and data forecasting techniques. Shell is currently working with IBM and DreamWorks Hollywood to achieve the data visualization. Although the analyses are done in the cloud (AWS), the visualizations are immediately available to the crew working at the local factory.
With the advent of Analytics in Shell, they have been able to achieve an upliftment of 6%-8% in the production in terms of money (in Billions)


Data-Driven Oil Fields

The search for new hydrocarbon deposits demands a huge amount of materials, manpower and logistics. With drilling a deepwater oil well often costing over $100 million, no one wants to be looking in the wrong place.

Surveying of potential sites involves monitoring the low frequency seismic waves that move through the earth below us due to tectonic activity. Probes are put into the earth at the spot being surveyed, which will register if the pattern of the waves is distorted as they pass through oil or gas. Shell uses fibre optic cables (created in a special partnership with Hewlett-Packard) for these sensors, and data is transferred to its private servers (maintained by Amazon Web Services). This gives a far more accurate image of what lies beneath. Data from any prospective oil field can then be compared alongside that from thousands of others around the world, to enable geologists to make more accurate recommendations about where to drill.


Asset Preventative maintenance optimization

Various sources and forms of data, including basic historical maintenance records and other information, can help operators detect and diagnose the root causes of poor performance and limit unplanned downtime of their machinery. By minimizing disruptions to production, or eliminating them before they happen, operators can maximize maintenance resources and keep production on schedule to meet financial goals

On some days, Shell assets generate no cash due to unexpected events and turnarounds overruns resulting in maintenance cost increase over the last few years. Old equipment that’s been in the field a long time can break down and ruin the best plans for maximized recovery.

Shell uses predictive analysis to identify the most effective set of maintenance activities and to create a means to decide which activities can be safely re prioritised. Machinery used in drilling has to operate in harsh conditions for prolonged periods of time so is prone to wear and damage. To counteract this, the machinery is fitted with sensors collecting data about its performance and comparing it with aggregated data, meaning parts can be replaced in an efficient manner and downtime minimized, further reducing overheads.

These are the few examples of what has been achieved, but there are speculations that there is so much more yet to be discovered with the help of this data. The possibilities are endless..!



No comments:

Post a Comment