I worked in refinery CDU's and I completely understand the issue you are talking about. On many occasions, I noticed the following:
1) The procedure for steaming of columns was really not developed, neither tested, nor improved based on actual field observations. Operators tend to introduce steam and then open all vents and drains, and keep it going that way forever until somebody says "hey, maybe we should stop and do the gas test". It is a terrible way of doing things but it happens almost everywhere. If this is the way Operations perform steaming in your refinery, maybe you can talk to the plant supervisor and try to put a decent steaming procedure as a start.
2) Draining of liquid inventory is the key for reduced timeline. Column bottoms, pump suction, draw off trays, and any other pocket where liquid can accumulate - you need to remove the inventory as much as possible. Otherwise it will take ages to vaporize all that liquid. The problem is usually with chimney trays where design leak rates are very small.
3) Look at the schematics of the isolated column and associated pipework and locate the vent and drain valves which should be open, and keep the other ones closed. Reduce the section volume as much as possible. If the section is very long, you won't see significant velocities at the pressurized end of the system. This further prolongs duration of steaming.
4) In some instances, it may be worthwhile to consider circulating hot water through the dirtiest section (column bottom/pump/hot preheat train/furnace/column) to get rid of sticky, viscous crude residue.
Dejan IVANOVIC
Process Engineer, MSChE