Machine learning can allow oil and gas companies to make better use of the enormous amounts of data as they try to maintain their pipelines. Last January, a major oil and gas company ran routine inspections of its thousands of miles of pipeline, using the same basic robotic device—the pig—that the industry has used for decades. However, this time, instead of sending data from the pig to a roomful of analysts and waiting months for results, the company applied a solution based on machine learning and data visualization.
The more time we spend with our clients the more we learn they are reliant on spreadsheets. To this point, we’ve jokingly considered changing our mission from “Predict pipeline failures, save lives and protect the environment… with the assistance of Machine Learning” to “We eliminate legacy Microsoft Excel spreadsheets”. Of course, all joking aside, we want to enable integrity management teams to spend their time doing high-value engineering. The latest Microsoft Excel sheet we have replaced is a customer’s version of a Crack Fatigue Analysis based on the modified Ln-Sec equation for cracks in pressurized pipelines.
During our time in the Microsoft Accelerator, Data Science, and Machine Learning cohort, we interviewed a few folks working in integrity management for pipeline operators to ask them to describe some of their most difficult challenges. We anticipated it would range from dealing with silos of data to spatially integrating risk data. However, we were surprised to learn that they simply wanted a solution that would accurately align features (welds, anomalies, valves, etc.
If you have ever looked through 20 years of inline inspection tally sheets, you will understand why it takes a machine learning technique (e.g. random forest, Bayesian methods) to ingest and normalize them into a database effectively. It would be a monumental task if attempted manually by a human … not to mention the risk of endless errors. However, by training a machine learning data classifier on enough log data, this task becomes the perfect scenario where data science can drastically improve integrity management practices.
It seems that every time an article or presentation is published highlighting some fancy new improvement to oil & gas pipeline integrity management related to inline inspection (ILI) and external data analysis, it comes attached with professional services to operationalize it. This is not so surprising, since historically the industry has been conditioned to compartmentalize everything on a project basis. It’s simply the way things have always been. Outside of data analysis this process probably makes sense.
