The Profile Query function is an important tool for finding 'funny' rail profiles. You can search for profiles that meet exception conditions (e.g. head loss > 50%), for transitions (i.e. sudden large changes in a measured value) or for 'spikes' (i.e. measurements that are very different from those of neighboring profiles on both sides). There are two kinds of 'funny' profiles: ones that are wrong and ones that are right.
'Funny' profiles that are wrong should be a tiny percentage of your data. The most common ones are profiles that have been identified as the wrong weight of rail. If you know the correct rail type, you can correct them at this stage, either singly or in batch. Other types of 'funny' profiles may be impossible to correct. They should be deleted.
When cleaning up 'funny' profiles, take care not to delete ones that are right! Data review and editing is an opportunity to discover rail problems that otherwise might remain undetected until they cause serious damage. You will quickly learn to distinguish bogus data from results that may be alarming but accurate.
The Track Condition Chart is another tool for reviewing data quality, as well as a useful report on its own. Run a chart for percent head loss, vertical wear, and rail type, looking for spikes and other unusual features. It may be useful to compare the run to an earlier one that has already been reviewed. Watch out for isolated profiles of a different rail type from their neighbors. These profiles may or may not be misidentified - sometimes, new 132 lb. rail is used as a plug to repair a section of worn 136 lb. If you are in doubt about a profile, you can view it by double-clicking on the chart. From the Profile Window, you can edit or delete the profile if appropriate.