Digital transformation/Late-life optimization: Harnessing data-driven strategies for late-life optimization

As the global oil and gas industry continues to mature, the number of fields in the late-life phase of production has steadily increased. Currently, more than 70% of the world’s oil and gas production comes via aging infrastructure built over mature reserves that are rapidly approaching their economic limit.1 

As reservoirs age, the natural pressure that facilitates hydrocarbon extraction diminishes, inevitably leading to reduced production rates and the associated economic implications. Extracting remaining reserves becomes increasingly challenging. To address this, the industry is increasingly turning to advanced technologies. Techniques such as polymer and chemical injection, gas injection, and thermal methods aim to maximize recovery rates and extend the productive life of these fields. However, these solutions often come with substantial upfront costs, requiring careful financial planning and investment. 

Aging infrastructure presents another hurdle. Pipelines, wells and other facilities are exposed to harsh environments, in addition to extreme operating conditions. Wear-and-tear of equipment over time leads to increased inspection and maintenance needs. The integrity of these assets becomes crucial, as any failure can result in significant environmental and safety risks. Maintenance and upgrading of facilities are imperative to ensure safety, operational efficiency, and compliance with advancing environmental standards. Addressing the maintenance requirements of aging infrastructure poses not only technical challenges but also financial and logistical complexities for oil and gas operators, with many assets in remote locations offshore and not easily accessible.¹  

Balancing the equation becomes a tricky task, requiring E&P operators to seek innovative solutions to maintain profitability. The financial viability of late-life fields becomes especially critical in the context of a global push towards renewable energy sources, forcing companies to weigh up their investments against the backdrop of a changing energy landscape. 

However, within these challenges lie opportunities for transformative change. Technologies that enhance the efficiency of late-life field operations, reduce emissions, and minimize environmental impact are becoming increasingly pivotal. Digitalization emerges as a key player in this evolution. Advanced data analytics, artificial intelligence, and the Internet of Things (IoT) empower operators to optimize production, predict equipment failures, and streamline maintenance processes. The digital transformation of mature fields not only extends their productive lifespan but also enhances safety and environmental stewardship.  

From upstream to downstream, for the energy industry to truly harness the power of data, a robust data management and integrity philosophy is critical to this success. This commitment ensures that the vast amounts of data generated are not only utilized for operational excellence but also adhere to the highest standards of security, integrity, and ethical use, underlining the industry's path toward a more sustainable and technologically advanced future. 

Data management for enhanced reliability. Managing data in mature assets poses a unique set of challenges stemming from the complex interplay of aging infrastructure, disparate systems, and evolving regulatory requirements that have changed in the decades since their inception. As mature fields navigate the later stages of production, maintaining data integrity is a critical requirement. Data integrity refers to both static and stored data, and its reliability and accuracy as it flows through interconnected processes, systems or components.  

As these assets reach the later stages of their lives, the sheer volume of historical data accumulated over years becomes a substantial hurdle. Legacy systems, often outdated and incompatible with modern technology, hinder seamless data integration and accessibility. Additionally, the diversity of data sources, ranging from conditioning monitoring to equipment maintenance records, presents a challenge in consolidating and analyzing information cohesively. 

The need to retrofit existing facilities for compliance with evolving environmental regulations further complicates data management efforts, demanding a difficult balance between historical data preservation and the integration of new technologies. Navigating this intricate landscape requires the sector to invest in robust data management strategies and embrace digitalization, ensuring that insights derived from mature oil and gas assets contribute meaningfully to operational efficiency and environmental sustainability.  

Aging infrastructure and equipment also can lead to sensor malfunctions or outdated instrumentation, introducing errors in data collection. Furthermore, different systems implemented at varying points in the asset's history often result in compatibility issues that can compromise the seamless flow and integration of data. Establishing robust protocols for data validation and safety, regular audits, and the implementation of advanced technologies is imperative.  

Assuring this reliability means ensuring data remains high-quality, consistent and timely. However, the vast amount and speed in which data are generated today can make data integrity more complex. This challenge is also not limited to the amount of accumulated data, but also to ensuring its accuracy and relevance over time.  

Extracting and collating data in late-life assets. Imrandd is a data science and engineering business that specializes in delivering data-driven insights and solutions for maintaining asset performance throughout the entire life cycle. The company was approached recently by a North Sea operator to support the development of late-life strategies for safety-critical pressure systems on their assets. The scope of work required the business to conduct analysis of integrity and inspection data to create optimized integrity and inspection plans, targeting essential tasks only before reaching cessation of production (CoP) dates. It was critical to prioritize safety-critical tasks, with technical justification in the most cost-effective way possible.  

The project’s focus was to maximize the best use of historical data to identify equipment, where inspection requirements could be removed or modified. In the initial stage of the project, an extensive review of the available integrity and inspection data for pressure systems across all installations was undertaken. This included an appraisal of data available, allocations for risk, probability of failure and integrity status, historical inspections and current planned intervals and dates.  

Assessments were conducted on both internal and external damage mechanisms for piping and vessel components with data retrieval first conducted to gather inspection reports, including internal and external integrity assessment data. This ensured that all legacy data were extracted to establish a clear picture of each asset and its component history. Similar data formatting was required, so full cleansing and correcting of data sources were delivered to the operator, to provide a singular source of data. 

Fig. 1. It is essential that data are analyzed to ensure their accuracy, and that gaps are identified early.

The next step involved the determination of metrics across the quantity and quality of data available, Fig 1. Throughout this phase, a substantial number of improvements were made to the data, which identified gaps and allowed the ability to address obvious errors and discrepancies. Inspection data were then pulled separately to provide a view on remaining life and support the Risk Based Inspection (RBI) process. This phase successfully identified circuits and vessels, where current inspection requirements could be challenged prior to CoP. These “quick wins” included finding 25 piping circuits on one asset with a low-risk probability of failure. This Improved the short- and long-term inspection workload across the asset, as inspection of these circuits was pushed out to a later date.  

Data extraction was carried out by Imrandd’s proprietary EXTRACT software across multiple platforms. The software ingests and distils disparately stored, effectively inaccessible data, then digitizes and consolidates it into one centralized database, ready for analysis. It utilizes a combination of techniques, including the most recent advances in Optical Character Recognition (OCR) and computer vision to enable vast volumes of data to be processed fast.  

The software produced more than 52,000 test points of data, dating back approximately 20 years. The extraction process was able to support the identification of systems providing an opportunity for optimization, which was crucial for the analysis and interpretation in phase 2 of the project. Continual checks and balances were provided, which demonstrated an estimated 20%-to-30% reduction in inspection scopes for phase 1.

Fig. 2. EXACT inspection coverage—data were plotted against remaining life and time since last inspection.

Innovative software applications. During phase 2, Imrandd deployed its proprietary analytics tool EXACT to accurately interpret the data it had collated, Fig. 2. The software cleanses, corrects and interprets large data sets, then maps and predicts equipment degradation to deliver actionable insights to significantly reduce OPEX and improve asset integrity management and plant reliability. The extracted data were run through a series of conditioning steps for input into the EXACT software. Once prepared, the tool was utilized to identify “good data” and “extreme data.” 

Good data have been collected, cleansed, quality-checked and screened as suitable for trending. Segregated or “extreme data” have been collected, cleansed and quality-checked and have been identified as an outlier that might potentially pose an immediate integrity threat or give cause for further investigation, usually conducted via desktop review by a discipline engineer. The output provided remaining lifespans of equipment and pipework while identifying circuits and equipment items where inspection intervals could be extended, or removed, from future inspection requirements, Fig. 3. 

Fig. 3. EXACT effectively plotted available inspection data, showing distribution of wall loss across corrosion circuits.

The detailed analysis optimized inspection scopes, recommending an approximately 30%-to-50% inspection reduction across all assets. To date, Imrandd has justified removing up to 50% of inspection efforts across the late-life North Sea assets. The result of the work and recommendations is expected to deliver additional savings and will provide the basis for further reduction during phase 3, which will define appropriate minimum allowable wall thickness (MAWT) for use in the remaining life analysis, which is projected to bring yet further optimization and removal of inspections scopes. 

Reducing inspections and enhancing safety. In a further demonstration of harnessing data for optimized inspection strategies, Imrandd was contracted by a separate global operator to deliver data integrity management, to enable better inspection planning and scheduling, without compromising on the safety of the assets in the short term. In support of growing sustainability objectives, a reduction in carbon footprint and ESG savings were also key objectives. The scope included cleansing and rationalizing topsides piping systems and associated inspection data before implementing the analytics process and using the results for optimization. 

Although a large amount of inspection data existed, there was a backlog in analysis of this data, which resulted in concern that some short-term integrity threats would not have been addressed soon enough. To mitigate the concerns over the volume of data, a staged approach was proposed. There was an increased risk of equipment failure and unplanned shutdowns during the planning phase, so the operator requested that analytics gathered and performed encompassed both internal and external asset conditions.  

Fig. 4. Data from over 500 external inspection reports were extracted and analyzed.

The company’s data team collaborated with the operator to prioritize systems and pipework for the first phase, with focus given to those due for inspection within two years. A series of additional analyses to interpret and further visualize the findings of the analytics was then conducted, with models built to accommodate external condition information, Fig. 4. By applying analytics and robust engineering methods to the cleansed data, the company’s experienced engineers transformed the planning and scheduling of inspection and maintenance up to CoP, stated for 2030, giving the operator confidence that they were following a plan that was based on analyzed, demonstrable insights. 

This analytics scope examined and trended data from a total of 3,500 lines in 182 corrosion circuits. This included more than 95,000 wall thickness measurements arising from inspections for internal corrosion. Data from over 500 external inspection reports were also extracted and analyzed. The optimized strategies resulted in a substantial reduction in the amount of inspection to be performed up until CoP. This varied by asset, but in total resulted in a 26% reduction in effort and spend on inspection related to piping. This represents a cost-saving of £2.65 million. The project also delivered significant sustainability benefits with an average saving of 26 tons of CO2 per asset on a yearly basis. Further optimization, as new data become available is expected to lead to additional cost-savings.  

Seven of the assets were completed as part of the full integrity management contract. This project has delivered a sound basis for the accurate scheduling of inspection activity in the short term and has paved the way for phase two, which will complete the analytics and give recommendations up to CoP in 2030.  

Supporting a data-driven future. Balancing the challenges posed by late-life production requires innovation, and the digital evolution of mature fields emerges as a key driver for the industry. Advanced technologies, such as data analytics, artificial intelligence, and the IoT, offer a lifeline for operators seeking to optimize production, predict equipment failures, and streamline maintenance processes. 

Legacy systems, disparate data sources, and evolving regulatory requirements demand a balance between preserving historical data and integrating cutting-edge technologies. However, recent examples underscore the tangible impact of utilizing data for optimized inspection strategies, offering a roadmap for cost-savings, sustainability benefits, and enhanced mechanical integrity. 

As we look to the future, the significance of robust data management strategies becomes increasingly apparent. The emphasis on data integrity, from static stored data to real-time sensor readings, is paramount in enabling data-driven decision-making. The challenges are vast, but the opportunities for positive change are equally immense.  

As the oil and gas industry grapples with the complexities of late-life production, it is clear that the road ahead must be paved with innovative technologies, strategic investments, and a steadfast commitment to data excellence. Only through such concerted efforts can the industry not only overcome its challenges but also emerge stronger and more resilient in the face of a rapidly transforming energy landscape. 

REFERENCE 

1. https://www.lowyinstitute.org/the-interpreter/great-offshore-decommissioning 

About the Authors

Steven Saunders

Imrandd

Steven Saunders is global head of new business at Imrandd. He has more than 26 years of industry experience, starting his career as a project engineer. He has held several asset assurance, senior inspection, lead integrity and asset integrity manager positions at Lloyds Register, Shell and BG Group. Prior to joining Imrandd, he was principal consultant—asset integrity management lead at Risktec Solutions, part of TUV Rhineland, based in the Middle East.