Data analytics can transform planned maintenance into predictive scheduling

Alex Endress, World Oil

One of the worst culprits that causes unplanned downtime on today’s drilling rigs is unplanned maintenance, to repair or replace equipment that has failed.

In efforts to reduce equipment failures, many drilling contractors have instituted meticulous, planned maintenance schedules, to ensure that critical equipment maintains only the highest level of acceptable operational integrity. However, there are issues with planned maintenance schedules—they can’t always ensure that equipment won’t fail, and increasing planned maintenance, to compensate for this dynamic, can be costly.

For instance, scheduling maintenance or grading the operational life of equipment on a general level—the average time it takes for a tool to wear out and need servicing or replacement under normal conditions—doesn’t take into account abnormal stresses that equipment endures. These abnormal stresses can cut down the life of a piece of equipment substantially, making it more prone to failure before its next planned maintenance appointment. It only takes one incident of equipment failure to cause significant downtime, or even a catastrophic accident. Yet, formulating a universal maintenance schedule for all equipment of a certain type is much easier said than done.

Enter Big Data analytics and predictive maintenance. The solution to these problems may be in the algorithms that oilfield service companies are embedding within software, to run on today’s computerized drilling equipment. It is now possible to track real-time operational data that are synthesized by software to tell the rig manager exactly what condition a tool is in, and how the operations that have been performed with the tool have changed its condition. Thus, the company will know when a piece of equipment is at risk of failure ahead of time. Conversely, the company can know exactly how much mileage it can get out of a piece of equipment that has been used sparingly.

Instead of planning maintenance schedules manually with the assumption that all equipment is working in its optimal state, as long as it has received regular maintenance according to corporate guidelines, data analytics can automate this process. Drilling contractors have started to realize this, and several have inked service contracts to improve maintenance predictability, using data-driven solutions during the past 12 months. In March, Noble Corp. announced that it would be using a digital optimization tool for operational efficiency, using data analytics on four of the company’s drilling rigs, with the prediction that opex costs could be reduced 20%, due in part to redefined maintenance strategies. In late 2016, Maersk Drilling announced similar projects and expectations for several of its jackups, semisubmersibles and drillships.   

As the concrete results of these projects become available, more drilling contractors, as well as E&P companies, will surely take notice. If, indeed, these companies see the desired opex savings, along with improvement in operational safety onboard the rigs, it will be tough for other companies to pass on the potential of digitization. If predictive maintenance via data-driven automation can completely eliminate equipment failure, safety, alone, should be a reason for operators to demand it on the rigs they hire. If, however, these digital solutions miss the mark, and do not perform as advertised, it may take some time before the industry at large trusts these technologies onboard its assets again.

The Authors ///

Alex Endress

Comments ///


{{ error }}
{{ comment.comment.Name }} • {{ comment.timeAgo }}
{{ comment.comment.Text }}