Progress in Mathematical Programming

Abstract : Most of the progress reported at the conference was on the theoretical side. Several new polynomial algorithms for linear programming were presented. The common feature to most of the new polynomial algorithms is the path-following aspect. The method of McCormick-Sofer for convex programming also follows a path. Efforts in the theoretical analysis of algorithms was also reported. Of special interest, although not in the main direction discussed at the conference, was the report by Rinaldi on the practical solution of some large traveling salesman problems. At the time of the conference it was still not clear weather the new algorithms developed since Karmarkar's algorithm would replace the simplex method in practice. Alan Hoffman presented results on conditions under which linear programming problems can be solved by greedy algorithms. In other presentations, Fourer-Gay-Kernighan presented a programming language (AMPL) for mathematical programming, David Gay presented graphic illustrations of the performance of Karmarkar's algorithm, and James Ho discussed embedding of linear programming in commonly used spreadsheets.