What software are you using?
If SKM, it doesn't have bus duct in the library for use as a bus - as-in calculating incident energy at that bus. It does have feeder bus for use as just that and it has an impedance component.
Are you trying to calculate IE for the bus as if the bus itself is what you are exposing yourself to live parts on or are you exposing yourself to the live parts at the disconnects or both? Is this plug-in bus that can have new switches plugged into while energized? The reason that the difference may or may not be important is, there is no enclosure or box per se while plugging in a new switch. There is of course the disconnect switch enclosure itself for exposure to live parts inside the switch.
The next question is, are you labeling the just the busway or just the disconnects or only the worst case between the two?
To model the busway you'll have to just add a bus, or segments of bus with a feeder bus of whatever amps between segments as jmoore suggests. Then you'll have to customize the bus gap and enclosure type and bus configuration (PNL, SWG, CBL, AIR, MCC) and (VOA, VCB, VCBB, HCB) To model each disconnect you'll have to add very short segments of feeder bus to additional busses representing the line side of each disconnect and customize the enclosure size and bus configuration again.
That is a lot of work to go to that extent and I expect your results between the bus plug and inside the disconnect will be similar if your bus gaps are similar. The biggest difference would be if you choose to model the bus plug in points as in AIR or in an enclosure. Here are a couple of screenshots from SKM to hopefully help describe what I am talking about.
Attachment:
BUSWAY SEG2.png [ 82.02 KiB | Viewed 21716 times ]
Attachment:
BUSWAY SEG2 arc.png [ 84.07 KiB | Viewed 21716 times ]
Attachment:
BUSPLUG1.png [ 76.82 KiB | Viewed 21716 times ]