Originally Posted by rsisk
Keep in mind that there are many ways the investigation by FIRST can turn out:
1. They are certain the cause of the problem(s) were found and fixed 100%. Ideally this is what we are all hoping for, but chances are probably 50/50 it will occur.
2. They are confident problem(s) were found and fixed. Issues don't recur in test environment but failures do appear at IRI (for example).
3. Things were found that could be the cause of the problem(s) but due to the intermittent nature of the problem or the inability to recreate the problem in a test environment no one can be sure the problem is solved.
4. Problem(s) are caused by multiple issues combined to produce a failure. Some of the defects are solved and the problem does not reappear... until later when another combination of issues causes the failure to recur.
5. Nothing is found
The nature of intermittent problems in a complex system means you will not likely be 100% confident problem(s) have resolved. The best you can hope for is moving the confidence level more towards the 100% level.
By FIRST taking the effort to conduct a significant dive into the problem, including attempting to recreate the environment as much as possible helps tremendously in letting us know that FIRST recognizes there is a problem and they are making an effort to resolve it. Results in the 1-4 category above will help move the confidence level in the field towards the 100% level (some more than others). If category 5 happens, well, the confidence level will probably stay where it is at.
Kudos to FIRST and all involved in the investigation for taking the time, effort, resources, and dollars necessary to deal with this issue.
This is a great way to describe the problem in front of us. The fact that First is going "all out" to reproduce the problem will help improve the chances of a positive outcome.