Nonlinear programming without a penalty function2002-01-010 PRE AI DOI:10.1007/s101070100244OriginalOriginal request for helpShareSaveAbstract En