Customer has this setup automatically-creating-dead-letter-resources.

It works well with anycast queue. But with multicast queue(or topic) if having multiple durable subscribers than all failed messages are sent to a common DLQ. It is difficult to group these messages and retry them again for the available durable subscriber.
Not everytime durable subscriber will be ready and available thus customer has to browse each message and than retry them. This is a tedious job for their operation team as they have very high traffic.

If we have an extra column with value of header _AMQ_ORIG_QUEUE than it will be easy to group messages(as than it wouldn't be required to browse each message)and retry them.