ARIS - SHARE YOUR IDEAS
We have written some custom, scheduled reports to check models for errors on a daily basis and then automatically correct simple errors (such as double spaces, wrong template used, non-default size objects, missing model prefixes on names, etc).
It would be great if an automatic error correction feature could be built into the core product that allows us to configure and pick from a set of defined error types and have them be automatically corrected, saving our modellers time and saving CoE time on performing quality checks.
If errors can be prevented in the first place, then that would be even better.
However, even with training, semantic checks and method filters, some errors continue to persist. This is why the retrospective correction of those errors is necessary.
Thanks
Cameron
Do you want the users to create incorrect models in the first place or should that be prevented?
Regards,
Philip
Thank you Philip.
The solution that we have built runs the corrections in two different scenarios. One runs as a scheduled report that runs overnight every night. One is embedded in our custom APG workflow just before the model review is performed.
Some models are in development for weeks or months before the user requests approval, so we run the fixes overnight every night as we want to minimise how long errors persist in the models.
In our APG workflow, we have the option to reject models, which sends the model back to the requestor to make necessary changes. Once they have made their changes, the model comes back to our reviewers to re-review it. Therefore we run the automated error fixes just before the review task, to ensure we are always covered.
I hope this all makes sense.
Thanks
Cameron
Hi Cameron,
thank you for sharing your idea. When do you expect these automatic corrections to be executed? Would it be sufficient if they get triggered right after the user requests approval?
Regards,
Philip