[Dip] Achieving master problem feasibility

[kim]

Hi

Is the method for achieving master problem feasibility described 
anywhere? From what I can tell, looking at the code and the master 
problems being generated, there's a two phase approach where:
- during the first phase DIP minimizes infeasibility until enough 
columns have been generated to ensure feasibility.
- this is done by adding positive and negative slack to all constraints.
- once feasibility is achieved the algorithm switches to the real 
restricted master problem without the "soft" constraints.

Is my understanding correct and is there anything else going on that is 
important when trying to achieve feasibility?

- Kim