I am in complete agreement with everyone who has essentially said, "Ignore to 80%/100% hysteria, calculate the demand load, size the conductor for the demand load, and size the breaker for the demand load." If using a 100% breaker, the continuous loads are calculated at 100 percent. If using a standard breaker, the continuous loads are calculated at 125 percent. If using optional calculations for dwelling units, schools, restaurants, whatever, use that demand load. There is no need to, nor is it Code-required to, multiply that value by 125 percent to offset the phantom 80% rating of a standard circuit breaker.
There is a benefit to 100% rated breaker assemblies insofar as when you use them, you can potentially reduce conductor sizes. Take, for example, a facility with 1200 amps of continuous load and 350 amps of noncontinuous load.
Using a standard circuit breaker, the conductors would need to be sized at 1200 * 1.25 + 350 = 1850 amps. The next standard size OCPD is 2000 amps, so you would need to increase conductors up to 2000 amps. Use 6 sets of #400 kcmil copper 75 degree C conductors or whatever else you want to get 2000 amps.
With the exact same load profile but using a 100% rated breaker assembly, the conductors would need to be sized at 1200 + 350 = 1550 amps. The next standard size OCPD is 1600 amps, so you would only need 1600 amp conductors to feed this identical load just by changing the type of breaker. So instead of using 6 sets of #400 kcmil copper 75 degree C conductors you can now use 5 sets of #400 kcmil copper 75 degree C conductors.
Now whether the savings afforded by reducing one set of conductors will offset the additional cost of a 100% rated breaker assembly is debatable.
Note that I kept referring to a 100% rated breaker assembly. It is not only the breaker that needs to be 100% rated; it is the entire assembly. As such, you can not simply purchase a 100% rated breaker, slap it in a standard piece of equipment, and apply that breaker at 100 percent.
