Predicting Software Reliability

[Alex Hu]

My group tests new software products and we want to use bugs "discovered" and "fixed" per week or any applicable time unit to measure software reliability (expressed in number of bugs?), relate this value to the average number of test cases executed and to predict reliability level at a later time in the test. Where should I start? Are there tools available?
Thanks!

[RS Support]

In general "Reliability Growth" models/ methodologies can be used. There is an example on Weibull.com you may find useful as a starting point at:
http://www.reliasoft.com/RG/examples/rgex5/index.htm