In addition, when testing

subsystems it is important to realize that interaction failure modes may

Software Design

not be generated until the subsystems are integrated into the total

system. A software reliability model indicates the form of a random process that defines the behavior of software failures to time. As an example, the failure data presented in the previous example will now be categorized into specific failure modes and types as shown in Table 3.

Evaluation of the delayed

corrective actions is provided by projected reliability values. The

demonstrated reliability is based on the actual current system

performance and estimates the system reliability due to corrective

actions incorporated during testing. The projected reliability is based

on the impact of the delayed fixes that will be incorporated at the end

of the test or between test phases. This field is populated with the value you select in the Date list on the Select Data Fields screen when you create an analysis based on dates. This field is populated automatically with the value you select in the Installation Date list on the Select Data Fields screen in the Reliability Growth Builder when you create an analysis based on dates.

Generally,

if the reliability of the failure mode meets the expectations of

management, then no corrective actions would be expected. If the

reliability of the failure mode is below expectations, the management

Environmental factors analysis and comparison affecting software reliability in development of multi-release software

strategy would generally call for the implementation of a corrective

action. This value is mapped from a query or dataset or manually entered when you create the analysis and is required.

The following table provides an alphabetical list and description of the fields that exist for the Reliability Growth Analysis family, which appear when you manually enter data on the Reliability Growth Analysis window. A Growth Model record stores information about the Reliability Growth model used to develop a Reliability Growth Analysis. The following table provides an alphabetical list and description of the fields that exist for the Growth Model family, which appear by default on the Growth Model datasheet.

When system failures are identified, the underlying flaws that are generating these failures are corrected, and the system’s dependability should improve through system testing and debugging. The conceptual reliability growth model must next be converted into a mathematical model in order to forecast dependability. These are non-homogeneous Poisson processes with Weibull intensity functions. Separate models can accommodate various phases of a reliability growth program. Reliability growth models are mathematical models used to predict the reliability of a system over time.

These models help the manager in deciding how much efforts should be devoted to testing. The objective of the project manager is to test and debug the system until the required level of reliability is reached. Reliability growth

Models

analysis is the process of collecting, modeling, analyzing and

interpreting data from the reliability growth development test program

(development testing). In addition, reliability growth analysis can be

• If you extrapolate the analysis results based on failure dates, this value is set automatically to True.
• Where “T” is the test time, “T0” is the time at the beginning of the monitoring period (initial time interval), “MTBFC” is the cumulative MTBF at time “T”, “MTBFI” is the instantaneous MTBF at time “T”, and “α” is the growth rate.
• The corrective actions for the BC-modes influence the growth in the system reliability during the test.
• This value is optional, but in order for dates to be displayed throughout the analysis, ALL failures must have a failure date.
• Evaluation of the delayed

  corrective actions is provided by projected reliability values.

done for data collected from the field (fielded systems). Fielded

systems also includes the ability to analyze data of complex repairable

systems. Depending on the metric(s) of interest and the data collection

method, different models can be utilized (or developed) to analyze the

growth processes.

A constant failure rate l can be expected on the assumption of a constant operating profile. The reliability function appears similar to the one shown above for hardware failures. This value is mapped from a query or dataset or manually entered when you create the analysis.

If you delete a value in the Date field for an event-based analysis that is based on dates, when you select Finish, an error message will appear. This field is populated with the value you select in the Cumulative Operating Time list on the Select Data Fields screen when you create an analysis based on cumulative operating time. In the second step, the individual  failures are entered into Table 2 of the calculator. The failure occurrence time is entered into the “Time” column, and the failure mode number to which the failure applies is entered into the “Failure Mode” column. In the final step, the total test time is entered into the appropriate field and the “Calculate” button is pressed.

Popper’s falsifiability criteria cannot be used in reliability growth models. This is mostly owing to the variety of models and parameters, making it virtually difficult not to find a model that fits with some given experimental or field data. A reliability growth model is a simulation of how system dependability evolves overtime throughout the testing process.

Several authors have suggested the use of the non‐homogeneous Poisson process to assess the reliability growth of software and to predict their failure behaviour. Inference procedures considered by these authors have been Bayesian in nature. Compares the performance of this model with Bayes empirical‐Bayes models and a time series model. The model developed is realistic, easy to use, and gives a better prediction of reliability of a software. Besides, testing environments may vary in practice due to factors such as hiring new testing personnel, the replacement of hardware, and the changes in testing tools or strategies, which result in changes in the efficiency of debugging.

Corresponds with the value selected in the Time Units list on the Select Data Fields screen for the analysis. This value is mapped from a query or dataset or manually entered when you create the analysis, and is required. Data record stores information about the Reliability Growth Analysis, which is a data format used to create a Reliability Growth Analysis.

This field is populated automatically with the value you enter in the Measurement Name box on the Select Data Format screen when you create an analysis. After you create an analysis, you can change the value in the Measurement Name field via the Select Growth Data Format window. If this value is False, the data is event count (e.g., number of failures). This field is populated depending on the value you select in the Measurement Data Format section of the Select Data Format screen when you create the analysis. Note that when testing and

assessing against a product’s specifications, the test environment must

be consistent with the specified environmental conditions under which

the product specifications are defined.

If you choose to extrapolate based on time or there is no extrapolation at all, this value is set to False. This field is only used for analyses based on cumulative operating time. This value is populated automatically with the value in the Start Time section of the Set Analysis Period window. This value is populated with the value in the End Time section of the Set Analysis Period window. We can try to pull ourselves out of these binds by our own bootstraps. Every so-called Reliability Growth Model (RGM) is based on certain assumptions about how failure rates vary as a result of fault elimination.

The test time necessary to grow the reliability from 500 to 2,000 hours can be calculated by substituting the values provided in Table 1 into the Duane model equations above and solving for “T”. If 4 test articles are used, then the total test time per article is 3,833 hours. The “Duane Method” calculator in the Quanterion Automated Reliability Toolkit – Enhanced Reliability (QuART-ER) (Figure 1) and QuART-PRO can be used to perform the calculations. If the required test time is prohibitive, then a more aggressive approach to precipitating and correcting failures should be considered, which could justify a higher growth rate. Reliability growth models are designed to forecast software behavior based on prior experience. In this scenario, previous experience is dependent on historical data; predictions cannot be validated by trials.

Kuei-Chen Chiu is currently an assistant professor in the Department of Finance at Shih Chien University (Kaohsiung Campus), Taiwan. And Ph.D. degrees from Industrial and Information Management of National Cheng Kung University. Her research interests include software reliability, https://www.globalcloudteam.com/ operations management, human factor, human resource management, and performance assessment. Related papers have appeared in such professional journals as Reliability Engineering and System Safety, Software Quality Journal, Journal of Taiwan Issue Economics and others.