Definition
A test rig is not a test environment of your own systems. It is an imitation of the other side. The distinction matters: in your own test environment you check that your code does what you think it does, whereas a rig lets you check that it does the right thing when the counterparty answers something unexpected.
It is that second half which is hard to arrange without a rig. Against a real system you can rarely order up a rejection, an expiring deadline or a message arriving out of order. Against a simulation you can provoke them deliberately, and repeat them identically every time, which is the precondition for a test being usable to measure a change.