What Is a Reliability System Model?

As a whole, industrial facilities typically have a difficult time forecasting system performance with confidence, which often results in reactive maintenance and wasted spending. Forecasting system availability using Quantitative Reliability Optimization (QRO) helps facilities balance performance targets, HSE risks, and the costs of managing both effectively.

Ask any facility owner, knowing where to make budget changes that will help facilities effectively manage their risks is challenging. Facilities often struggle with:

  1. Determining potential failure events at their facility
  2. The impact of those unplanned events
  3. The prescriptive maintenance tasks or actions that should be taken to prevent those events and improve overall performance in the most cost-effective manner and at the optimal point in time

QRO creates a causal link between every data point, task, or potential change in relation to the performance of each asset and, by extension, overall facility performance.

How Does Forecasting System Availability Differ from Conventional Models?

While some conventional reliability models such as Reliability Availability Maintainability (RAM) modeling are useful in approximating availability (typically in the design phase), these types of models are often limited in scope, are static and use generic industry data.

This means that once the facility is operating, the model becomes out of date and is not accurate. Forecasting system availability combines builds upon the first two elements of QRO – Asset Risk Analysis (ARA) and Lifetime Variability Curve (LVC) – using first principles engineering analysis and data science to predict the health of each asset and connects this information across an entire system and ultimately the facility.

Leveraging the mathematical relationships built from the reliability system model, facilities can connect the probability of failure for individual assets, and calculate their expected availability over time. Each individual asset is then incorporated into a system reliability model to show the impact that each individual asset has on the entire facility.

Facility performance can then be forecasted by the direct and dynamic correlation of how data on each asset impacts facility performance. This enables facility owners to see when there are likely to be performance issues and where to focus their efforts to maintain facility reliability.

System model in Newton™

System model for a unit in Newton™

System model for a facility in Newton™

Benefits of Forecasting System Availability

By using QRO to forecast system availability, facilities are able to:

  • Model system reliability configuration, including system redundancy, bypasses, slowdowns, and other operations parameters
  • Prioritize tasks and activities that will have the greatest impact on facility performance
  • Understand reliability vulnerabilities and forecast facility performance based on real-time data

To learn more about how forecasting system availability combines elements of an ARA and an LVC to drive more holistic system models, read about QRO here.

More resources like this