Currently we are working on a project to estimate software reliability. We want to consider the uncertainty factors in our calculation. That's why we are trying to find out those factors. If you have anything in your mind, please suggest.
I introduce this paper for evaluating uncertainty factors maybe affect software reliability.
https://www.researchgate.net/publication/223109402_An_analysis_of_factors_affecting_software_reliability
https://www.researchgate.net/publication/289593169_Qualitative_Evaluation_of_Software_Reliability_Considering_Many_Uncertain_Factors
THE MAJOR UNCERTAINTY FACTORS AFFECTING IN
SOFTWARE COST ESTIMATION, G Rajkumar et al , Int.J.Computer Technology & Applications,Vol 4 (3),419-424.
Extrinsic Influence Factors in Software Reliability:
A Study of 200,000 Windows Machines
Christian Bird, Venkatesh-Prasad Ranganath, Thomas Zimmermann, Nachiappan Nagappan, Andreas Zeller.
Variation in Operational profile is an important uncertainty factor
First, you need a reliability metric (or grid of metrics) of some sort and then empirical data about it or about other measures that may correlate with it.
One thing that has caught my eye is non-feature-request bug reports, the rate of arrivals versus repairs over time, the bugs which have significant user pile-ons and that remain open for extended periods of time. Also watch out for won't fix and bugs closed because users don't do the verification work.
In this case, I think we're looking at reliability of the production and maintenance of the software, and indirectly reliability of the software itself.
There are still problems about qualifying reliability. Defect reports about loss or corruption of data are clearly more serious, as well as loss of effort, and then there are others that determine a software product unfit for purpose depending on specific circumstances and use cases. That may be very difficult to get to.
So what are the qualities and measures that one is attempting to sample/predict?
Dear Dennis Hamilton , reliability is itself related with probability. Now adding uncertainty with it is really very hard to predict.
Internal Quality: the totality of the attributes of a product,
in terms of its constitutional and design aspects, which
determine its ability to satisfy defined and implicit needs
when used under certain specific conditions.
External Quality: the extent to which a product meets its
explicit and implicit needs when used under certain
specific conditions.
Quality in Use: the extent to which a product used by one
specific user meets its needs to achieve
specific objectives with effectiveness, productivity and satisfaction
in a particular context of use.
Check the software by using the extremes of the parameters that it was designed to run within. In other words push the software to the limits to make it fail. Secondly try repeatbility. If you see a failure, rerun the program until you are able to reproduce the same error. Thirdly, when errors are happenign collect the fequency of such occurrences.
Dear Md Asraful Haque,
Another factor which can affect the software reliability is the frequency of requirements changes. Also, technology changes, devices failure, political instability, behavior of users, and so on could be affected the reliability as uncertain factors.
I think for considering some of uncertain factors, the defined method should be acted using hybrid calculations (fuzzy & stochastic).
Dear Md Asraful Haque ,
Please follow these articles:
[1] An analysis of factors affecting software reliability.
- January 2000,Journal of Systems and Software 50(1):43-56
- DOI: 10.1016/S0164-1212(99)00075-8, Source: DBLP
[2] Software Reliability Qualitative Evaluation Based on Modified Delphi Hierarchy Process
- 3rd International Conference on Mechatronics, Robotics and Automation (ICMRA 2015)
- 2015. Published by Atlantis Press
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling