Why the Different CBC Experimental Design Approaches?

There are 5 ways to build experimental designs in Lighthouse Studio’s CBC software: Balanced Overlap (default), Complete Enumeration, Shortcut, Random, and imported (customized) from a .CSV file. What are the goals of the design algorithms? Why do some algorithms allow levels to repeat within the same choice task (level overlap)? Which approaches do a better job of capturing preferences for the respondent who doesn’t want to compromise on must-have or must-avoid levels? Which approaches are better for capturing interaction effects?