What is Reliability Prediction?

Reliability Prediction is one of the most common techniques used by engineers when evaluating product and system reliability. By taking into account all parts of an electro-mechanical system with their associated operating and rated parameters, Reliability Prediction analysis enables you to predict your system’s failure rate or MTBF (Mean Time Between Failures). Reliability predictions tend to be utilized most often in a product’s design stage. However, they can be used throughout the product lifecycle to provide insight into the reliability of your product and help target areas for potential improvement.

To perform the failure rate estimation, prediction standards that define the equations used to quantify reliability are utilized. Some of the most commonly used and widely accepted reliability prediction standards include MIL-HDBK-217, Telcordia, 217Plus, and ANSI/VITA. The equations delineated in the standards were built by analyzing a huge amount of field data over a long period of time. Statistical analysis was then used to determine which ones best modeled the failure characteristics of the accumulated data. Depending upon your industry’s needs, one standard may be the most applicable or you may use a combination of standards. Our in-depth guide on reliability prediction standards provides detailed information on the differences between the various documents to help determine which is most applicable for your use.

What Data is Required for Reliability Prediction?

The inputs required for reliability prediction analysis are the components, or parts, in your system and their operating parameters – for example, environmental conditions, electrical stress, and temperature stress. The output is the predicted failure rate of the system expressed as the number of failures over a particular time period. The most commonly used units are Failures per Million Hours (FPMH) and Failure per Billion Hours (FITs). For example, a failure rate of 10 FPMH indicates that the system is estimated to fail 10 times over one million hours of operation. Because systems can be vastly different in terms of composition and use, there is no defined cross-industry failure rate specification. However, in most cases, there is a target failure rate that is considered to be acceptable. Reliability Prediction analysis can help you achieve that target and can aid in product development to ensure new designs achieve or surpass previous baselines.

Because reliability prediction analysis takes into account all the component parts that make up your system, it provides a well-grounded estimate of predicted failure rate. However, for complex systems, this can mean working with a mountain of input data! This article provides details on efficient management strategies for your reliability prediction data so your analysis is both easy and effective.

How to Streamline your Reliability Prediction Analyses

Making Reliability Prediction Data Entry Efficient

Rather than performing calculations by hand, software packages are often used to assist in reliability prediction analysis. These can reduce potential calculation errors and significantly decrease the time it takes to calculate your system’s failure rate. There are several key features to look for to make your reliability predictions most efficient.

Component Libraries

A Component Library is a database that contains a list of parts from known manufacturers. For each part, the parts library contains applicable component data required for reliability prediction calculations. Choosing a software package with a thorough built-in parts library can help eliminate the often-time-consuming step of defining each part’s electro-mechanical information.

Relyence Reliability Prediction’s supplied Parts Libraries contain hundreds of thousands of components and have built-in support for both the NPRD (Non-Electronic Parts Reliability Data) and EPRD (Electronic Parts Reliability Data) databases. In addition, you have the ability to create a custom part library with your own list of known parts and associated part data.

Intelligent Part Mapping^TM

Unique to Relyence Reliability Prediction, Intelligent Part Mapping provides an additional means for retrieving a part’s electro-mechanical parameters. Given your input in the part’s Description field, Intelligent Part Mapping extracts as much relevant component data as possible.

For example, Relyence will automatically recognize the description “Cap Cer 10uF 50V” as a ceramic capacitor with a capacitance of 10uF and a rated voltage of 50V. By this small example, it is clear how Intelligent Part Mapping greatly improves the efficiency of component data entry for reliability prediction analyses.

BOM Importing

Oftentimes, the best place to begin a reliability prediction analysis is with your BOM (Bill of Materials). Because the BOM contains a list of all parts that make up your system, importing this information into your analysis is an efficient way to begin entering your part data. However, BOMs typically only include a portion of the data needed for reliability prediction analysis, such as part number and description. Therefore, combining BOM importing with other data entry aids, such as component libraries and Intelligent Part Mapping, offers a more complete method for data entry.

In Relyence Reliability Prediction, you can directly import your BOM from Excel. Relyence’s step-by-step Import Wizard guides you through the import process and leverages both Parts Library look up and Intelligent Part Mapping capabilities to obtain any additional component data that is required for calculation.

Making Reliability Prediction Analyses Efficient

In addition to streamlining the data entry process, there are a number of additional high-powered features available that can have a significant impact on the efficiency of your reliability prediction analysis tasks.

Reusing Reliability Prediction Data Across Analyses

When new products are developed, they may include subassemblies from previous designs. In addition, subassemblies are often used across product lines for design and production efficiency. For example, the same PCB (printed circuit board) could be used by an automotive manufacturer in their cruise control system across all vehicles. In these situations, it can be helpful to have a master subassembly repository to allow for easy data reusability across reliability prediction analyses.

In Relyence Reliability Prediction, this can be accomplished by using our unique Knowledge Bank^TM feature. An easy way to understand the Relyence Knowledge Bank feature is by example. For a sample case, let’s assume we have a reliability prediction analysis completed for a quadcopter drone. Now, we are developing our next-gen product: an octocopter drone.

How to Use the Relyence Reliability Prediction Knowledge Bank

The quadcopter drone includes subsystems for the Motherboard, GPS, and Ground Controller. We want to reuse the GPS subsystem in our new drone. Using Relyence, we can easily add the GPS to our Knowledge Bank and use it for both product analyses:

1. Add the GPS subsystem from the Quadcopter Drone analysis to your Knowledge Bank.
   1. From the Properties pane of the GPS, click Bank > Add to Knowledge Bank from the toolbar.
   2. Optionally, after you get a message that says the GPS was successfully added, you can click Bank > Go to Knowledge Bank to see the GPS subsystem in your Knowledge Bank. (Note: You can also access the Knowledge Bank by clicking Libraries > Knowledge Bank in the sidebar.) This will serve as the Master copy of your GPS subassembly with all parts and part data included.

2. Add the GPS to your Octocopter Drone analysis.
   1. In your Octocopter Drone analysis, click Edit > Edit Tree above the tree hierarchy.
   2. Click the Search button in the top toolbar, type “GPS” into the Search field and click Search. (Note: You can pick either to insert as a Sibling or Child subsystem depending on where the GPS subsystem fits.)
   3. Select the checkbox next to the GPS subassembly and click Insert.
   4. The GPS subassembly will be inserted into the octocopter drone analysis.

In addition to the obvious advantage of reusability, the Relyence Knowledge Bank also includes automatic data synchronization through its Push mechanism. This allows you to keep any changes made to the master copy in sync across all analyses.

Mixing Models for Effective Reliability Prediction Analyses

The choice of which reliability prediction standard to employ can be contractually required depending on your industry. In cases where it is not, it can be beneficial to choose multiple models as certain types of parts and environments may be included only in specific standards. For example, one popular way of performing reliability prediction analysis is to use MIL-HDBK-217, Telcordia, and the NPRD/EPRD databases together. In this case, you get coverage of almost all device types used in product design.

In Relyence Reliability Prediction, you can set the calculation model to use on each part or each subassembly, allowing you to use a combination of models for the most effective analysis.

Integrating with other RAMS tools

As a standalone tool, reliability prediction analysis is a vital tool for MTBF assessment. However, you can leverage that power for even greater benefit by integrating with other reliability analysis tools. Ways to use an integrated approach include:

• Use Failure Rates calculated in Reliability Prediction in MIL-STD-1629A FMECA studies to more accurately target areas for continuous improvement efforts.
• Integrate Reliability Prediction with your FRACAS (Failure Reporting, Analysis, and Corrective Action System) or CAPA (Corrective and Preventive Actions) to compare calculated to actual field failure rates.
• Link your calculated Reliability Prediction failure rates to RBD (Reliability Block Diagram) to account for system redundancies.
• Use your calculated Reliability Prediction failure rates in Fault Tree Analysis (FTA) to better understand system failure and risk reduction strategies.
• Integrate Reliability Prediction analysis with MIL-HDBK-472 Maintainability studies to help optimize your repair and maintenance activities.

Relyence Reliability Prediction is part of the Relyence Studio. The Studio platform enables you to create a customized reliability tool set package for cross-module capabilities suited to your needs. For more information about reliability tools and methodologies, download our comprehensive “An Overview of Reliability Analysis Tools and Techniques” white paper.

Conclusion

Reliability Prediction is a powerful tool for ensuring your product reliability goals are being met and in supporting your continuous product improvement efforts. By taking into account all components that make up a system, reliability prediction analysis provides an accurate and comprehensive failure rate estimation. Get the most out of your reliability predictions by leveraging techniques to streamline data entry, improve calculation effectiveness, enhance data reusability, and integrate with other RAMS tools.

Relyence Reliability Prediction provides an efficient platform for performing reliability predictions and MTBF analysis. If you are interested in learning more about Relyence Reliability Prediction or our integrated Relyence Studio platform sign up today for your own no-hassle free trial. Or, feel free to contact us to discuss your needs or schedule a personal demo.