Our Approach

As this was a brand new asset, no historic data was available, so our data scientists used a variety of techniques to create dynamic predictive models of the entire gas compression system.

Upon start-up, these models allowed us to monitor thousands of data relationships on behalf of the customer on a continuous basis, so that our team could alert the asset to any threats to stable operations at the earliest possible opportunity.

Actionable Insights

By helping this customer to predict and anticipate a number of potential gas compression system trips, they were able to make adjustments that avoided millions of dollars of production loss, as well as significantly reducing maintenance man hours and expenditure.

Insights

During the first 12 months of application we helped this customer to achieve:

$681k

reactive maintenance cost savings

400+

maintenance manhours saved

17

gas compression system trips avoided

867

kboe production loss avoided

A practical example of an insight on this system

Dry Gas Seals

Our data analysts observed that there were significant differences in the time taken for the compressor’s dry gas seals to arrive at normal operating conditions following a start-up.

Working with the asset support team it was discovered that at one point in the start sequence, a valve should be automatically operated. However, this valve was sticking, or had to be operated in manual. The variation in start times came from the time between pressing the start button and when the operator manually operated the valve.

This discovery led to the customer going back to the OEM and revising the machine’s logic to remove the dependency on the operator remembering to open the valve.