Midstream operators in the U.S. lose $10 billion annually to pipeline leaks and theft, according to Deloitte. Given that breaches can result in expensive cleanups along with environmental damage and stiff penalties for regulatory noncompliance, it’s easy to see how losses could rapidly pile up.

And with the slimmer margins associated with sluggish oil prices, operators can’t afford to lose valuable product for any reason. Beyond these day-to-day challenges, midstream as a business is becoming more complex. Operators are transporting a wider variety of petrochemical products to an increasingly diverse number of destinations.

Products and routes can also change frequently, driven by market demand and pricing.

Equipment is mission-critical, as assets like combustion engines, compressors and pumps bear the weight of keeping product flowing and avoiding catastrophic system failure. Under often extreme conditions, these complex devices require vigilant monitoring and diligent upkeep to sustain production.

Underused data

Digital technology in the form of sensors, device connectivity and data collection have been widely deployed by all major players across the industry to address environmental and safety issues. However, even though insight-rich streams of equipment data are at operators’ fingertips, digital information remains woefully underutilized.

Several factors contribute to this oversight. The overwhelming speed and volume of data generated by production assets and control systems; the wide, remote distribution of operations; and the high cost of network transport and storage from these far-flung assets are among the top inhibitors. As a result, many midstream companies severely restrict the amount of data they use for analysis—often using only 1% to 2% of the total—leaving extensive potential benefits on the table. Optimizing maintenance and repair of capital equipment is one such missed opportunity.

The Industrial Internet of Things (IIoT) can advance digital transformation initiatives by helping operators streamline maintenance workflows and realize greater business benefit from data they’re already collecting.

Maintenance models

Unplanned downtime—besides interrupting the profitable flow of product—often requires expensive, time-consuming emergency repairs in remote locations and exposes operators to environmental, regulatory compliance and safety risks.

Maintenance of critical assets is crucial in avoiding unplanned downtime, but it’s also one of the most challenging operational aspects of the midstream operator’s business. Improperly executed asset upkeep can lead to an unexpected failure that halts production, threatens lives and the environment and incurs major emergency repair expenses. This sort of reactive maintenance is risky and doesn’t support profitable business operations.

On the other hand, over-maintaining equipment can create needless expenses if the repairs aren’t necessary. This imprecise approach to preventative maintenance is familiar to anyone who has ever changed oil in a car at prescribed mileage intervals. Operators risk incurring excessive and redundant costs by servicing assets on a similarly fixed schedule based on specific metrics—such as elapsed time or cycles of operation—regardless of the actual condition of the machinery.

To move toward a more exact, proactive approach, operators should consider condition-based maintenance (CBM). This helps control costs and optimize equipment performance by servicing assets based on specific factors known to affect performance.

For example, using real-time monitoring, an IIoT solution can evaluate a variety of factor-based data from sensors—like fluctuations in temperature, pressure or flow rates—and alert operators when certain conditions are met so they can take action. It is also possible to tie solutions into enterprise operational or maintenance applications to coordinate parts and service personnel.

Providing context
However, while sensors can provide mountains of information, without further context that data can create a flurry of false alerts and alarms that can actually cost operators more in unnecessary service calls.

Imagine a scenario in which there is a section of pipeline hundreds of miles from the nearest service team. Sensors are flagging the operations center that the pipeline’s temperature is excessive and rising rapidly. A crew is quickly dispatched, only to find a maintenance team busy steam-cleaning that section of pipe.

In this scenario, time-based rules and rules covering multiple conditions are necessary to detect which changes in the equipment’s state truly warrant action. By further tying into other enterprise systems, such as the application used to schedule repair technicians, an IIoT solution can provide much greater visibility into how changes in the state of equipment relate to broader operational activities.

Operators seeking to adopt more proactive maintenance models can utilize IIoT to effectively forecast future failure events. By applying analytics to historical repair and maintenance data, as well as current operating conditions of the equipment and any external factors that are influencing behavioral changes, an IIoT solution makes it possible to employ predictive maintenance practices.

With this, operators can determine the root cause of a pending failure and what parts need repair or replacement, and execute on detailed repair plans set out by the system. As a result, service teams can arrive at the work site with the correct parts and repair instructions needed to perform the necessary work quickly, and get it right the first time.

How IIoT helps

Advances in software that performs the heavy lifting of complex data analysis have made it possible for IIoT solutions to extract powerful insights from all equipment data, control systems and relevant enterprise systems. Digital twins are just one application for these advances. Utilizing information from a population of similar assets along with machine learning and sophisticated analytics, an IIoT system can create digital models of equipment.

Operators can use these digital twins to run time-based scenarios and queries that can predict future equipment states and potential failures. Subject matter experts from the operator’s organization guide the development and fine-tuning of rules and responses to incidents identified by the system. Based on this feedback, the system becomes progressively smarter and over time can automatically perform actions to mitigate possible failures or safety issues.

But this still leaves the dual problems of network connectivity with remote assets and storage of massive amounts of data—much of which is simply data indicating normal conditions. These issues can be solved with a hybrid approach that uses a combination of cloud and edge analytics, tapping processing power that is likely already available on connected equipment to perform analytics locally to determine what data can be discarded and what gets forwarded for further processing in the cloud. In this way, operators can collect and manage data from vast populations of devices at disparate sites and sift through everything in order to optimize maintenance and repair processes.

The result is greater accuracy without the cost burdens of transmitting and storing massive amounts of ultimately irrelevant data.

Keeping profits flowing

Midstream operators can achieve greater productivity and virtually eliminate unplanned downtime by adopting an IIoT solution that can effectively use all their connected equipment data in conjunction with existing operational and enterprise applications. By looking at their maintenance and repair operations holistically and moving toward a more proactive maintenance model, operators can streamline maintenance and repair workflows while realizing greater overall benefit from data they’re already collecting.

A well-designed IIoT solution—developed with guidance from operators’ subject matter experts—can harness data to better predict asset failure, diagnose root causes more effectively, troubleshoot issues more efficiently, automate corrective actions for safety and continued operation, and maintain equipment based on actual condition and environmental factors. And by implementing a system that uses both edge and cloud-based analytics, operators can overcome connectivity and bandwidth obstacles to implement these enhancements across large asset populations in remote locations.

The result is an IIoT solution that delivers tangible benefits for dramatically better business operations.

Dave McCarthy is senior director of products at Bsquare Corp.