Introduction
Over the past ten years, an integrated, international energy company experienced a significant drop in plant-wide availability at one of its larger refineries. Plant leadership identified the hydrocracker unit as one of the primary contributors to the drop in availability and focused efforts on improving that unit. The plan, which had a goal to maximize the availability of the hydrocracker, included a series of asset management improvement initiatives, capital upgrades, and performance improvement plans.
After the initiatives were completed, plant leadership remained uncertain that these activities would actually help them achieve their availability goals. While the plant had implemented a risk-based inspection (RBI) program for its fixed equipment and a reliability centered maintenance (RCM) study for its critical machinery, these efforts seemed subjective and overly conservative, and provided a static view of equipment reliability. As a result, these methodologies were not capable of sufficiently quantifying results that would provide plant leadership with the confidence that the availability improvements they were seeking would be realized.
Plant leadership asked: “Should we be doing more? Are we spending too much? Can we be certain that the actions we are taking are worth the investment? How can we be more confident that the planned maintenance, monitoring and repair, replace, and upgrade activities are worth the investment and will ensure a step change in availability?”
Quantitative Reliability Optimization
The plant needed a solution that would help them better evaluate equipment risk and predict future availability, so plant leadership decided to pilot Quantitative Reliability Optimization (QRO). QRO is a data-driven methodology that combines the best traditional reliability methodologies with data science principles and subject matter expertise (SME), enabling plants to drive and improve complex reliability decision-making. This approach blends the risk assessment of both fixed and non-fixed assets into a single model, removes data silos, and provides plants with insights to reduce unplanned downtime, increase safety, and improve spending performance with statistically supported confidence. Just as the industry evolved its approach to assessing risk with methodologies like RBI and RCM, QRO is the next advancement of reliability modeling.
QRO provides four major benefits to facilities:
- The ability to predict future availability by leveraging existing data
- Accurate forecasting of probability of failure (POF) and consequence of failure (COF) for both fixed and non-fixed assets in the same methodology
- Facilities with limited data can use industry analytics and subject matter expertise to build and inform the data models for facilities
- The ability to update predictive models in real-time with live data connections including process, monitoring, work order, and task data, which allows risk and mitigation plans to remain evergreened
Comments and Discussion
There are no comments yet.
Add a Comment
Please log in or register to participate in comments and discussions.