217Plus Calculations

The 217Plus standard consists of two basic steps: like other standards, it uses component models to estimate each component’s failure rate and sums those rates to yield an initial system failure rate. It then takes system-level factors into account to calculate a final failure rate for the system.

The system-level calculations use the following types of data:

• Optional Data - Calculated from Pi Factor values applied to the system.
• Predecessor Data - Calculated from a specified predecessor predicted and observed failure rate.
• Test/Field Data - Calculated from the test/field data set of the system, which represents a series of failure datapoints for the system.

Note that in this context, a "system" can be either a standard system component, or block components underneath the standard. You will use the Define Optional Data, Predecessor Data and Test/Field Data at this Level option to define the data at a specific level.

System-level calculations can be performed at only one level in a hierarchy. If you have a block or standard set to define the data used for these calculations, then no descendants underneath that block or standard can define the data as well. This approach ensures that predecessor data or Bayesian analysis with Test/Field data adjustments will be not applied multiple times to the same component.

If a parent is set to define system calculations, then any descendants that were previously set will automatically have the setting disabled and ignored.

With the set of the rules described above, a user can analyze systems where predecessor or test/field data of the new system are available not for the system/top assembly level but for lower level items (e.g., a subsystem). In such cases, system reliability engineering principles for series configuration are applied to calculate upper level blocks and top system reliability.

Note: If a component is on the same level as a block (Component – Block series relationship) with the Define Optional Data, Predecessor Data and Test/Field Data at this Level option enabled then the component has only the applied initial assessment of failure rate that is added to sum of failure rates of the blocks at the same level.

Neufelder Software Prediction Model component

The 217Plus standard includes a special component called the Neufelder Software Prediction Model that is intended to capture the failure rate of software components in the system. This component has no inherent failure rate but it can affect the system level calculations if the parent block or system has the Define Optional Data, Predecessor Data and Test/Field Data at this Level field set to Yes.

Note: The values of the Neufelder Model Assessment and % Severe Faults fields are set to constant values and cannot be changed. The Number of Months Until Next Release is limited to integer values from range 1 to 6. More information about Neufelder Software Reliability Prediction Model can be found at missionreadysoftware.com.