Hi Sasi,
I would be reluctant to follow this path. The availability check code really is not designed to be executed in the update task. However, outputs are executed in the update task. If the ATP check is executed via an output, this means that every time we might hit some error in the ATP check, we will encounter an update error (and there are many opportunities for error in ATP). Consequently, this means any other changes made to the delivery will be rolled back and lost. Also, by executing the ATP check in the update task, the system will not actually tell the user there is 0 available.
If you were to follow this, it would need to be tested very thoroughly.
Regards,
Noel