Predictive data of inner space: the past and future of well analytics

15 June 2017

By Derek Crombie, VP for software development at Lloyd's Register.

The petrophysics and well logging industry has been intrinsically driven by mass data for decades. However, what we can do now with the insight we extract from wells within the subsurface - or 'inner space' as I like to call it - is breaking new ground in terms of future well placement and drilling plan strategies.

Mainstream interest in the world of 'big data' and predictive analytics has spiked over recent years, but the concept is not such a giant leap for the oil and gas industry. Our ability to process data quickly has rapidly developed and the challenge now is to unearth the most valuable insight from ever-increasing data volumes.

It is 20 years since I started working with well and subsurface data software. In 1997, I was a junior petrophysicist who had just started working alongside the team that eventually released IP™. At the time, petrophysical interpretations took hours, even days; we used to input figures manually for each zone and painstakingly plot the impact of each modification on the screen. The industry was ripe for new technology, which is how IP™ - originally known as Interactive Petrophysics - was conceived.

That analysis can now be done in minutes. Two decades on, and IP™ has gone beyond the realms of pure petrophysics and is now utilised by geologists, geophysicists, reservoir and production engineers for the intuitive, interactive approach to their analysis and interpretation.

A new approach to predictive analytics

Today, the breadth of insight that we can mine has expanded massively and IP™ has grown with the changing dynamics of the industry. It has undergone significant upgrades since becoming part of the Lloyd's Register software portfolio, and taken predictive analysis to a new level using Domain Transfer Analysis (DTA).

Traditional log data prediction is often based on a statistical approach using preconceived rock and fluid models, which have limited scope and flexibility. With domain analytics however, we are able to navigate the volumes of drilling and log data from previously drilled wells in order to uncover predictions, including porosity, permeability, saturations, TOC and mineral volumes while drilling. This allows almost real-time decision making and insight to the identity and characteristic of the formation being drilled.

What makes Domain Transfer Analysis different from the plethora of statistical approaches is the fluid dynamics based mathematics that underpin it and our multi-dimensional technique. Using reverse modeling, it takes data from nearby wells, side-tracks and the well itself – to uncover a depth of insight that other approaches would miss.

It's a fascinating technique that can guide the sector through previously unchartered territory and provide insight that would have been buried in a mass of 'big logging data'.

Some say it is impossible to know what the future holds, but when it comes to 'inner space' data analytics can help predict the future.

Derek Crombie is the VP for software development at Lloyd's Register. He will be previewing the 2018 versions of our industry-leading IP™ well data and petrophysical interpretation software at the SPWLA Annual Symposium 2017, from 17 – 21 June in Oklahoma City. Find out more about LR's oil and gas software.