It depends (don't you just love that phrase?).
The plants I was involved in had large oversized dessicant beds so we could really run out into the plateau area of adsorption capacity versus time and typically, we had to shut down due to pressure drops across the bed rather than cycle times. Thus, the material in the beds could be reused after the fines causing the high dP were removed.
The molecular sieve was simply screened first to remove the large hold-down and support balls and then screened again to remove the fines. We simply contracted with a local company who came in with the necessary equipment and we unloaded the beds into their equipment. After screening, we made up the loss of dessicant fines with new material (I can't remember if we loaded the new dessicant on top to take advantage of its higher water capacity or on the bottom to take advantage of a shorter mass transfer zone, I think it went on the top).
What material are you drying with your alumina? You want to make sure there isn't any hazardous material on the dessicant, pyrophoric material, when the contractor is handling it. I'll see if I can track down some specifics on the procedure for you on the area of safety. I was more involved with this at arm's length than the actual planning and execution.