Entering 217Plus Software
When 217Plus is selected as the calculation model, selecting a part category of Software and a subcategory of 217Plus Software causes parameters to appear in the Prediction Data pane. The 217Plus parameters for a software component are very different from those for an electronic hardware component. Additionally, process grades are not used to adjust the failure rate calculated for a software component as they are for a hardware component.
 |
217Plus and PRISM are the only calculation models with a component model for software. When one of these models is selected, you can calculate the predicted failure rate of a software component. When another model is selected, you can use the general software reliability equation from The Rome Laboratory Reliability Engineer's Toolkit. For more information, see General Software Model.
|
The following table describes the part parameters required for a 217Plus software component. The default values are for when little information is available because the software component is assumed to be new and not widely deployed. As the software component becomes more widely deployed, more accurate values should be entered for Fault Latency and Fault Activation %.
|
Parameter
|
Type
|
Description
|
|---|---|---|
|
Months Since Deployment
|
Real
|
The number of months since the software was released.
|
|
Lines of Code (x1000)
|
Integer
|
The number of lines of code in the software. The unit is thousands of lines of code. Thus, 10 would be entered if there were 10,000 lines of code. There is no default. You must enter a value in this field.
|
|
CMM Level
|
Choice List
|
A value from the Capability Maturity Model (CMM) that indicates how mature the software is in regard to its development process. The CMM, developed by the SEI (Software Engineering Institute) is widely accepted for the evaluation of software reliability. Once the inherent fault density is estimated as a function of the achieved CMM level, it is converted to a failure rate based on the defined operational profile of the software.
|
|
Fault Density, Initial
|
Real
|
The number of faults per thousand lines of code at the software’s initial release. The default is based on the selected CMM Level.
|
|
Defect Stabilization Level
|
Real
|
The level at which the software failure rate stabilizes relative to the initial defect density. This percentage value indicates how much the reliability of the delivered code is improved after faults are identified because of processes that the maintaining organization have in place. If no reliability growth processes are in place, use 100 as the default stabilization level. The default is based on the selected CMM Level.
|
|
Fault Latency
|
Real
|
The average number of times a fault occurs before the fault is corrected. The default is 2.0
|
|
Months To Stabilize
|
Real
|
The number of months until code maturity (or fault stabilization) is reached. The default is 48, which is standard for the initial release. If this is a subsequent release, the value in this field should be changed. The standard for subsequent releases is 24.
|
|
Fault Activation %
|
Real
|
The percentage of modules exhibiting the occurrence of faults. The default is 100.
|
|
Average Severity
|
Real
|
The percentage of faults that is disruptive or critical to the customer. The default is 50.
|
|
[Derating Parameters]
|
---
|
For more information, see Derating Parameters.
|
|
|
Windchill Risk and Reliability uses the Neufelder model for software reliability prediction when 217Plus:2015 standard is selected. This standard is owned by Softrel, LLC (www.softrel.com). This model is described in the section 2.2.16 of the 217Plus:2015 standard. For more details, see the Windchill Risk and Reliability Reference Guide from the Reference Documents area on PTC website.
|