Karen Boman speaks with Prasantha Jayakody of Bsquare about the benefits of condition-based maintenance and the challenges of implementation in the oil and gas industry.
Graphic from Bsquare.
Reducing non-productive time related to equipment maintenance and breakdowns is necessary during the downturn as companies seek cost reductions. To achieve these goals, the industry is looking to shift from reactive or even preventative maintenance to data-enabled maintenance, which is automated and gathers information in real-time. Data-enabled maintenance includes predictive maintenance, predictive diagnostics, and condition-based maintenance (CBM).
CBM is a bit of a general catch-all term that encompasses several different data-driven maintain/repair concepts. It should be viewed as a “maturity model” as predictive maintenance can be done without CBM, but CBM can’t be done without adaptive diagnostics (AD), says Prasantha Jayakody, senior product manager for Bsquare Corp., an Internet of Things solution provider. CBM builds on AD, which gathers existing structured and unstructured data to create a comprehensive view of maintenance activities and recommend the best fix. CBM uses analytical software to determine equipment condition in real-time, based on equipment operating use and sensor data combined with the equipment’s AD history.
CBM is one of the tools oil and gas companies could use as they seek to digitalize their businesses. Currently, the industry is hoping to leverage digital tools to glean more insight from the massive amount of data gathered in operations (OE: June 2017).
One example of industry’s interest in CBM is drilling contractor Transocean and GE Oil & Gas’ announcement in January that GE would provide CBM and maintenance services for pressure control equipment on seven of Transocean’s rigs over the next 10-12 years. The agreement, signed in late 2016, leverages GE’s digital capabilities to shift from event and calendar-based maintenance to condition-based monitoring and maintenance. Working with GE on parts forecasting and service scheduling will allow Transocean to optimize operations by proactively planning and minimizing between-well maintenance, GE said at the time.
OE: How did the offshore oil and gas industry view CBM prior to the 2014 oil price downturn? How has the view changed post-downturn?
Prior to the 2014 oil price downturn, CBM was commonly viewed as a ‘nice-to-have’ and while most firms recognized the benefits of data-driven maintenance, few firms pursued it. With oil prices north of US$100/bbl, companies could afford to run to failure or practice interval based preventative maintenance without serious repercussions to the bottom line. In the years following the price downturn, operating budgets were slashed while companies focused on projects with reduced scope and the highest return on capital. With operating budgets greatly reduced, redundant roles were eliminated across maintenance and reliability teams and basic equipment maintenance became the status quo.
OE: What are some other factors spurring greater interest in the offshore oil and gas industry to consider the wider use of CBM?
Safety and reliability are some of the biggest forces driving CBM adoption across the industry. The environmental and financial impact of catastrophic failures is a top concern of management teams and are central to safe, efficient operations. The complexity of offshore facilities balanced by a streamlined workforce presents an opportunity to leverage data to make sure work is performed on the right equipment at the right time. Riding on top of these forces is the remote nature of offshore facilities. In the event of equipment failure, spare parts or intrinsic knowledge may be hundreds of miles away, so performing repairs should be done with the proper teams onsite during a time that will have the least amount of impact on ongoing operations.
OE: Why do you think that producers have failed to adopt the architecture they need? Will this prevent them from reaping the full benefits of CBM?
Data is not the issue; modern offshore facilities generate vast amounts of data from thousands and thousands of sensors. But, identifying which data is useful and what is noise is one barrier to adoption. (McKinsey reported that less than 1% of the sensor data on offshore rigs is used in the decision-making process). Another obstacle is managing this data and the costs associated with moving it to the cloud where advanced analytics and machine learning can be performed. With edge (fog) computing, much of this analysis can now be performed onsite avoiding costly cellular/satellite transmission fees while simultaneously reducing latency. Additionally, we’re starting to see organizational changes across teams and hybridization between operations technology, engineering technology and information technology. Small project teams are being assembled to address critical issues and are driving scalable change across companies.
OE: What kind of challenges must the industry address for CBM to significantly impact subsea production?
The benefits of CBM are magnified in subsea production, given the limited accessibility to equipment. However, companies need to up their CBM smarts to coordinate maintenance across subsea equipment – equipment needs to be grouped together and machine learning behind CBM analysis needs to optimize the maintenance interval across the entire group of equipment. Additionally, companies need to look beyond CBM and leverage predictive failure technology to decrease downtime of subsea equipment even further.
OE: What other challenges face the oil and gas industry in adopting CBM?
Scalability is a challenge for CBM for several different reasons. In modern rigs, data collection is natively built into most of the equipment, but it becomes more difficult in older installations. Granted, equipment doesn’t have to be replaced outright as data collection technology can be retrofitted to older equipment, but this retrofit still requires planning and capital. Secondly, the ideal scenario for CBM is when you can compare performance characteristics and operating parameters across several identical pieces of equipment. It’s not realistic to have a single make and model of a given type of equipment across all sites but some level of uniformity is required.
Prasantha Jayakody is a senior product manager for Bsquare. He has over 20 years’ experience in the software industry, working with companies across several industries. Prasantha graduated from the University of Pennsylvania with a BA in mathematics and a BS in computer science.