### Table 5: Number of edges cut ]Ec[ on refined airfoil problem ( IE[ =

1991

Cited by 267

### Table 3: More CirCut Results on Max-Cut problems from the Torus Set

2001

"... In PAGE 9: ...hree columns of the table, respectively. All the reported times are in seconds. From the table, it is abundantly clear that an average run of CirCut is much faster and produces better quality cuts on all four test problems. More results are reported in Table3 for CirCut using different values of N. These results indicate that the variations between the average and best cut values are quite moderate, and they also show that even with N = 0 (no further improvement attempted after minimization),... ..."

Cited by 23

### Table 2a. Problems Statistics After Default Cut Generation

1995

"... In PAGE 13: ... The last column gives the percentage of the gap closed due to presolve and cut generation. Table2 b shows the statistics for the instances where disjunctive cuts were necessary. There were four such cases, and these were the only four models for which disjunctive cuts were used.... In PAGE 17: ... Table2 b. Problems Statistics After Disjunctive Cut Generation Disj.... ..."

Cited by 12

### Table 2a. Problems Statistics After Default Cut Generation

"... In PAGE 9: ... The last column gives the percentage of the gap closed due to presolve and cut generation. Table2 b shows the statistics for the instances where disjunctive cuts were necessary. There were four such cases, and these were the only four models for which disjunctive cuts were used.... In PAGE 13: ... Table2 b. Problems Statistics After Disjunctive Cut Generation Disj.... ..."

### Table 1: Branch-and-cut on compact problems.

1996

"... In PAGE 14: ... In addition, some of the problems were run with the condition that if a variable is positive, it must be at least a certain input value (minimum transaction level). Table1 presents data for these runs. The column headed \rank quot; refers to the rank of the objec- tive, \ineqs quot; stands for the number of inequalities in the original formulation, \cols quot; the number of columns, \nodes quot; is the number of branch-and-cut nodes needed to solve the problem, and \cuts quot; is the number of disjunctive inequali- ties separated by the algorithm, counted when generated (i.... ..."

Cited by 27

### Table 1 Statistics for the torus set of MAX-CUT problems.

2001

"... In PAGE 10: ...3 for details). In Table1 , we give statistics for this set of problems; note that the sizes of the graphs are given as (|V |, |E|). In the table, the columns Lower bound and Upper bound give the best lower and upper bounds on the maximum cut value known to us to date, and the column SDP bound gives the SDP upper bounds on the maximum cut values.... ..."

Cited by 23

### Table 2. Max-Cut problems solved on 1 processor.

"... In PAGE 5: ...2 Results We report here some preliminary computational results. In Table2 , we show results on Max-Cut problems solved in serial. The first two columns tabulate the prob- lem name and the number of serious / total iterations (Ser/Tot).... ..."

### Table 15: Cuts per node in the branch and cut tree of problem test500 5 in table 8

1998

### Table 15: Cuts per node in the branch and cut tree of problem test500 5 in table 8

1998

### Table 3. 2-Matching Problems: cut-and-branch approach.

"... In PAGE 9: ...able 2. 2-Matching Problems: impact of the multiplier selection rules. MIP solver for concluding the optimization. This is illustrated in Table3 , where some large 2-matching instances are solved through ILOG-Cplex MIP solver with and without the addition of an initial pool of CG cuts generated by 100 calls to our separation procedure. 3.... ..."