Verdazo AI

Practical machine learning for Oil & Gas

Deep technical and domain expertise

We combine expertise in machine learning and analytics with real world Oil & Gas experience.

Industry’s fastest time to value

Our team delivers practical results you can act upon, in weeks rather than months.

Transparent, pragmatic approach

We identify and quantify the key factors that contribute to asset performance.

How Verdazo AI clients are using machine learning

Our solutions are applicable to several key areas in Oil & Gas. They’re
practical, understandable, and actionable.

Incorporate more data

Understand the impacts and importance of all your data.

  • Apply more data, analyses, and interpretations
  • Quantify the effect of all inputs on well performance
  • Get a more complete and strategic understanding of your assets

Explain well performance

Fortify your understanding and challenge your assumptions.

  • Understand feature importance and what drives production
  • Assess performance by production area and stratigraphic zone
  • Quantify all of these outputs to drive clear action

Usability and interpretability

Get usable, data-driven deliverables you can act on.

  • Test out hypotheses using an interactive predictive model
  • Explore visualizations that build model understanding and provide deeper insights
  • Access data and results with Verdazo BI or other visual analytics software

How we partner with you to deliver machine learning

We’re transparent, collaborative, and highly practical when working with our clients.
You’ll get results you can use and models you can trust.

Engage in a collaborative process

Our committed, expert team will work with you throughout the process and will continue to provide support as you require.

Deepen understanding of outcomes

Get improved, data-driven explanation of well performance, and an understanding of how better predictions can influence the way that you approach your assets.

Get usable, trusted, and interpretable results

Utilize predictive models to assess feature importance and test hypotheses, new datasets you can explore, and visualizations that bring it all to life.

Explore our industry-leading analytics content

May 31, 2019 by

Brian Emmerson delivering talk on machine learning for Darcy Partners in Houston

Verdazo Analytics’ Director of Data Science Brian Emmerson will deliver a talk on interpretable machine learning techniques in Houston, Texas on June 5th. The one-day forum is presented by Darcy Partners and is focused on analytics and the internet-of-things. Brian’s talk will focus on the ability of robust visualization techniques to investigate machine learning model predictions, with examples from North American unconventional resource plays. To learn more about the conference or to register, click here.

Read More
May 13, 2019 by

Verdazo Analytics at geoconvention 2019, Calgary

There are three ways to see Verdazo’s team in action at geoconvention 2019 in Calgary. Marcelo Guarido will be presenting Machine Learning Strategies to Perform Facies Classification Tuesday May 14th at 9:00 am in the Glenn Room 206. Verdazo Analytics’ Director of Data Science Brian Emmerson will co-present The Importance of Geoscience in using Machine Learning to Predict and Optimize Well Performance – Case Study from the Spirit River Formation Wednesday May 15th at 10:35 in the Telus Room 104-105. You can also catch members of the Verdazo team at Booth 503 to learn more about our visual analytics and machine learning capabilities.

Read More
May 12, 2019 by

Augmenting Well Production Analysis with Subsurface Data

The Montney Formation, located in the Western Canadian Sedimentary Basin, is developed in a multi-zone stack throughout the fairway.  Unfortunately, these refined target zones are not captured in public data.  For analysis, it is important to differentiate Montney Wells into multiple target zones because they vary significantly in reservoir properties both vertically and laterally. Identifying the target zone based on sequence stratigraphy is a valuable process but can be time-consuming. In this blog we show a quick method to differentiate target zones using a depth-based approach that is helpful when you have limited time and resources.  This workflow can be applied to any map-based data to derive a data set suitable for well production analysis. For contour-based geologic data to be useful for well production analysis, we need a map-derived value for each individual well.  To accomplish this, we started with a publicly available Geologic map of the Montney Top in Meters Subsea (BC OGC, 2012). Figure 1: Digitized and interpolated Montney Top Subsea TVD (True Vertical Depth) Structure Map, (OGC, 2012). The Montney Top Structure map contours were digitized (Figure 1), so that interpolated well-values could be derived.  Using point-sampling, the Montney Top Depth was extracted at the intersection...

Read More