Covariate Software Failure and Reliability Assessment Tool (C-SFRAT)
The Covariate Software Failure and Reliability Assessment Tool (C-SFRAT) is an open source application that applies covariate software reliability models to help guide model selection and test activity allocation. The primary functions of the C-SFRAT include:
- Displaying model fit and failure intensity plots of selected hazard function and covariate combinations
- Prediction of future failures and failure intensity based on a specified testing activity profile
- Comparison of fitted models based on information theoretic and predictive goodness-of-fit measures with user-defined weighting
- Recommendations for test activity allocation to maximize defect discovery within a specified budget or minimize the total testing resources required to discover a specified number of defects.
C-SFRAT is compatible with Windows, macOS, and Linux running Python 3.x.
Python can be installed from: https://www.python.org/downloads/.
The Python libraries required to run C-SFRAT are:
To install compatible versions of the libraries, first open a terminal/command prompt window and navigate to the C-SFRAT root directory. Then, run the command
pip install -r requirements.txt.
To run C-SFRAT, first navigate to the C-SFRAT root directory in a terminal/command prompt window. Run C-SFRAT using the command
python main.py. Depending on your Python installation, you may instead need to enter
Command Line Arguments
python main.py [-v] [-d]
- -v, --verbose Run application in verbose mode, offering more detailed information printed to the terminal
- -d, --debug Run application in debug mode, offering additional details that can help with debugging but are not required for typical use
The main folders of C-SFRAT are:
- core - contains mathematical functions used for model fitting and prediction
- datasets - contains example data sets
- models - contains hazard function definitions
- ui - contains graphical user interface layout definitions
The covariate model that C-SFRAT applies was presented in:
V. Nagaraju, C. Jayasinghe, and L. Fiondella, “Optimal test activity allocation for covariate software reliability and security models,” Journal of Systems and Software, p. 110643, 2020.
This material is based upon work supported by the National Science Foundation under Grant Number (#1749635).
This research was supported by (i) the Naval Air Systems Command (NAVAIR) through the Systems Engineering Research Center (SERC), a Department of Defense (DoD) University Affiliated Research Center (UARC) under Research Task 139: Software Reliability Modeling and (ii) the National Science Foundation under Grant Number (1526128).
Copyright and License
Code release under MIT LICENSE.