Data Analytics, News

Data Analytics learner project: Automating the analysis of oil field efficiency

01 April 2021

An engineer and a geologist from the energy company bp, have completed an automation project that has modernised their approach to reservoir management. 

The project, which will result in safer, more efficient oil well management, was delivered by Carwyn Adler and Stephen Richardson as part of their Corndel Data Analytics Diploma, using skills acquired on the programme. They explained why they decided to embark on this project and the impact that it's had.

The challenge

Collecting surveillance data is a crucial part of successful oil field management and development. In one of bp’s principal producing reservoirs in Iraq, a huge amount of data is collected and analysed to see if the production wells are working as they should be. A high level of water production is one of the primary causes of well failure and watercut data informs the team about the proportion of water to oil in a well. In 2019 alone, more than 3,000 watercut measurements were taken. About 90% of this data had to be manually collated and analysed. This manual approach brought with it risk of human error and blind spots around likely failures and damage. 

The project 

Through the ‘Watercut Automated Assessment Tool’ project, Carwyn Adler and Stephen Richardson automated most manual components of the current procedure. At the start of their Data Analytics Diploma, they learned SQL and immediately used that knowledge to gather oil well data from numerous databases across the Middle East Organisation. This involved collaborating with the team in Iraq who were painstakingly doing the tests by hand. 

As the course developed and honed their skills in Python, they quickly saw its potential in enabling them to incorporate an element of automated trend analysis to ensure accuracy, consistency and more objective interpretation of the data. They developed a complex multi-stage piece of code that would effectively automate the analysis of the data. They built flexibility in the script to enable it, in the future, to incorporate information from other reservoirs.  

Their programme tutor and coach, Sam Ashcroft, was on hand throughout to advise and give feedback on their suggestions and work.  

The outcome 

Carwyn and Stephen worked on this project as part of their Diploma, with exceptional results. They managed to turn a 30-hour manual data analysis job, carried out quarterly, into an automated process that takes 10 minutes. This automation means that the analysis can be done on a more frequent basis, allowing quicker responses to ad-hoc requests and more informed discussions. The quality of the analysis has significantly improved, and the risk of human error and bias has been eliminated. Not only can the team now better manage the wells in terms of efficacy; they are also more likely to identify if a well is going to fail. Being able to schedule repairs or address the issue, instead of paying hefty prices for emergency repair work, will unlock significant savings. 

What’s next? 

Carwyn and Stephen are proud of what they’ve achieved and are confident that further value will be derived from more advanced analytics capability.  

“In the past, I had attempted to develop my data science skillset through various online courses. It wasn’t until I signed up to the Corndel Data Analytics Diploma, with the requirement to apply the skills developed in the course on relevant work related projects, that I really began to understand and unlock the true potential of data science for my role. Time with Sam, our programme tutor, has enabled me to further push my skillset and get me past any tricky technical problems.” Carwyn Adler 

"The Corndel Data Science Apprenticeship has given me skills to take my work to the next level. Its taught me how to automate tasks I used to do (slowly or repetitively) in Excel, but also introduced me to new techniques which generate additional insights. The online classes are great and the support from our tutor Sam was excellent. The Apprenticeship provides a solid foundation in Python programming language and establishes this quite quickly – really within a few months. I’ve been able to apply Python to several projects at bp already, the main one in collaboration with my teammate Carwyn on the same course. Geology and Data Science are very complimentary – both categorise, process and present data into clear, visual products, and I think we’re in a quite exciting time where there are a lot of opportunities to combine them." Stephen Richardson

Data Analytics Diploma coach and data scientist, Sam Ashcroft comments, "The real-world business impact of this exceptional workplace project cannot be understated. As part of their Level 4 Data Analysis qualification, Carwyn and Stephen took the SQL and Python content from the course and implemented it in their department, to automate what was previously a costly and time-intensive process. In completing this data analysis project, they improved business forecasting models, reduced time-cost and human-error and saved dozens of hours each week which can now be spent automating other time-consuming processes in the business. I am pleased to have been the Professional Development Expert for two such intelligent and enthusiastic Data Analysts. Well done Stephen and Carwyn."

Share this with your network

Recent posts

You might also be interested in...

Stay connected with Corndel
Keep up-to-date with the latest information from the team, including all of our
must-read news and insights.