E. Balas, S. Ceria, G. Cornuejols & N. Natraj, "Gomory cuts revisited". Oper. Res. Lett., vol. 19, pp. 1--10, 1996.  Home/Search   Document Details and Download   Summary   Related Articles   Check

....bound, we have found that it is an e#ective method for perturbing x # , thus giving the separation heuristics a second chance to succeed. However, the Gomory cuts tend to be quite dense, so we only perform this trick at the root node of the branch and cut tree, and even then only once. As in [9], we generate an entire round of Gomory cuts rather than only one at a time. That is, for each j such that x # j is fractional, we find the associated row of the simplex tableau, which will be of the form: x j # i x i = x # j , 12) where NB represents the set of non basic variables. Then we ....

E. Balas, S. Ceria, G. Cornuejols & N. Natraj, "Gomory cuts revisited". Oper. Res. Lett., vol. 19, pp. 1--10, 1996.

....edges in the support graph of x . We have also investigated several families of facet de ning mod k cuts for both the symmetric and asymmetric TSP. Recent developments in cutting plane algorithms, such as the work of Balas, Ceria and Cornu ejols [2, 3] and Balas, Ceria, Cornu ejols and Natraj [4] on lift and project (disjunctive) cuts and Gomory cuts, put the emphasis on separation of large classes of inequalities which are not given explicitly. The approach developed in this paper provides still another tool for tackling hard problems. Future theoretical research should be devoted to ....

E. Balas, S. Ceria, G. Cornuejols, N. Natraj (1996). Gomory cuts revisited. Oper. Res. Lett. 19, 1-9.

....plane algorithms based on general purpose cuts such as Gomory or disjunctive cuts exhibit slow convergence. This can be alleviated somewhat by adding several cuts in one go before reoptimizing by dual simplex, see for example Balas, Ceria Cornu ejols [1] and Balas, Ceria, Cornu ejols Natraj [3]. In general, however, it is preferable to use inequalities which take problem structure into account, especially inequalities which are deep in the sense of inducing facets (or faces of high dimension) of the polyhedron P I (see Padberg Gr otschel [19] Nemhauser Wolsey [18] It is often ....

....cutting and branching within a single algorithm. This yields the branch and cut technique, in which cutting planes are used at each node of the branch and bound tree to strengthen the LP relaxations (see Padberg Rinaldi [21] Balas, Ceria Cornu ejols [2] Balas, Ceria, Cornu ejols Natraj [3]; Caprara Fischetti [6] In branch and cut it is normal to use inequalities which are valid for P I as cutting planes. Inequalities of this kind are valid globally, i.e. at every node of the branch and cut tree. Thus, any violated inequality generated at any node may be used to strengthen the ....

E. Balas, S. Ceria, G. Cornuejols & N. Natraj (1996) Gomory cuts revisited. Oper. Res. Lett. 19, 1-9.

....is to use general purpose cutting planes which do not require or exploit any a priori knowledge about the structure of the problem at hand. The Gomory mixed integer cuts (see Gomory [11] which have been successfully used in a branch and cut framework (see Balas, Ceria, Cornuejols and Natraj [4]) or the disjunctive cutting planes (see Balas [2] revisited and implemented as the equivalent lift and project cuts (see Balas, Ceria, Cornuejols [3] fall into this category. Another possibility is to generate special purpose cutting planes that are based on the polyhedral analysis of the ....

E. Balas, S. Ceria, G. Cornuejols and Natraj, "Gomory Cuts Revisited", Operations Research Letters 19 (1996), 1-9.

....the underlying structure. Examples are (mixed) integer Chv atal Gomory cuts or disjunctive cutting planes. The idea to derive disjunctive cutting planes is to split a polyhedron into two or more disjoint polyhedra, study these polyhedra individually and convexify the individual descriptions. In Balas, Ceria, and Cornu ejols [1993] 1996] it is shown that disjunctive cutting planes help solving general mixed integer programs. We introduce here a somewhat contrary approach. Instead of breaking an integer program apart we study the convex hull of integer points that lie in the intersection of several polyhedra simultaneously. These ....

....(12) reads x 1 x 2 Gamma 4 3 x 3 5. It is violated by the feasible solution x = 3; 3; 0) T . Similarly, w Gamma = Gamma1 and the inequality x 1 x 2 Gamma x 3 6 from (13) is violated by the feasible solution x = 13; 0; 6) T . The problem discussed in Example 4.4 is not new. Balas, Ceria, Cornu ejols, and Natraj [1996] give another example in conjunction with Chv atal Gomory cuts. Here the same difficulty arises, when Chv atal Gomory cuts are generated at some node, which is not the root note. When integer variables are involved these inequalities are usually only locally valid for this node and its ....

[Article contains additional citation context not shown here]

Balas, E., Ceria, S., Cornu'ejols, G., and Natraj, N. (1996). Gomory cuts revisited.

....algorithm for general integer linear optimization problems. The practical performance of Gomory s original algorithm is not good. However, recent work in computational integer programming has shown that Gomory s cutting planes may be very useful when used in a branch and cut algorithm, see e.g. (Balas, Ceria, Cornu ejols Natraj 1996). 3.2 Disjunctive cutting planes We brie y mention a second principle to derive cutting planes, which is based on disjunctive programming (Balas 1979, Sherali Shetty 1980) A disjunctive optimization problem is of the form maxfc T x j h2H A h x b h ; x 0g (16) for H nite, A h ....

....disjunctions of the form Ax b; x 0 x j d Ax b; x 0 x j d 1; 18) for some j 2 f1; ng and d 2 Z. A very successful branch and cut algorithm for general linear 0 1 optimization, which is based on disjunctive cutting planes, is the lift and project method developed by Balas, Ceria Cornu ejols (1993, 1996). 4 The elementary closure The set of vectors satisfying all Gomory Chv atal cutting planes for P is called the elementary closure P 0 of P . Thus, if P is de ned by the system Ax b, where A 2 Z m n and b 2 Z m , then P 0 is de ned as P 0 = 2R m 0 T A2Z n ( T Ax ....

Balas, E., Ceria, S., Cornuejols, G. & Natraj, N. R. (1996), `Gomory cuts revisited', Operations Research Letters 19.

....BMIPs that can be used successfully in branch and cut. Our inequalities are derived from the flow cover inequalities that were introduced by Padberg, Van Roy, and Wolsey [1985] and Van Roy and Wolsey [1986] and implemented by Van Roy and Wolsey [1987] Balas et al. 1993] use disjunctive cuts and Balas et al. 1996] use Gomory cuts to solve BMIPs by a branch and cut algorithm. Usually, flow cover inequalities are not facet defining and need to be lifted to obtain stronger inequalities. However, because of the sequential nature of the standard lifting techniques and the complexity of the optimization problems ....

E. Balas, S. Ceria, G. Cornuejols, and N. Natraj (1996). Gomory cuts revisited, Operations Research Letters, to appear.

....factorization. Finally, an additional problem with these cutting planes was that, if generated within a branch and bound tree, the cut was not valid throughout the tree, since the basis representation used to generate these cuts, assumed that certain variables were fixed. Recently work by Balas, et al. [1999] have suggested approaches to overcome the ill conditioning problem (by carefully considering when to branch and when to cut) Similarly, they have adopted lifting techniques originally derived for polyhedral based cuts to force the validity of the cuts throughout the tree. We will first introduce ....

E. Balas, G. Cornuejols, and N. Natraj (1999) "Gomory Cuts Revisited", Operations Research Letters, to appear 1999.

.... (1993a) Clique problems: Grotschel and Wakabayashi (1989, 1990) Coloring: Lee and Leung (1993b) Nemhauser and Park (1991) Covering, packing and partition: Balas and Padberg (1972) Padberg (1973,1977,1980) Nemhauser and Trotter (1974) Trotter (1975) Wolsey (1976b) Balas and Zemel (1977) Balas and Ho (1980), Balas and Ng (1989a,b) Cornu ejols and Sassano (1989) Laurent (1989) Nobili and Sassano (1989) Sassano (1989) Chopra and Rao (1993) Ferreira et al. 1996,1998) Muller and Schulz (1996) Cheng and Cunningham (1997) Cut polytopes: Barahona and Mahjoub (1986) Barahona et al. 1988) ....

E. Balas, S. Ceria, G. Cornu' ejols and Natraj (1996a) "Gomory cuts revisited", Operatiosn Reserch Letters 19 1--9.

....of nodes and the number of edges in the support graph of x . We have also investigated several families of facet defining mod k cuts for the TSP. Recent developments in cutting plane algorithms, such as the work of Balas, Ceria and Cornu ejols [3, 4] and Balas, Ceria, Cornu ejols and Natraj [5] on lift and project (disjunctive) cuts and Gomory cuts, as well as the recent work of Applegate, Bixby, Chv atal and Cook [2] on the separation of TSP cuts obtained through local analysis of the fractional point, put the emphasis on separation of large classes of inequalities which are not given ....

E. Balas, S. Ceria, G. Cornu'ejols, N. Natraj (1996). Gomory cuts revisited. Oper. Res. Lett. 19, 1--9.

....very small problem instances within reasonable CPU time. State of the art algorithms for mixed integer optimization preprocess the problem (see e.g. Sav94] and use a cutting plane algorithm in each node of the branch and bound tree in order to strengthen the relaxation (see e.g. BCC94b] [BCC94a], NSS94] We will implement some of these techniques in a future version of this solver for mixed integer optimization problems. ....

Egon Balas, Sebastian Ceria, and Gerard Cornuejols. Gomory cuts revisited. Technical report, Columbia University, 1994.

....and cutting plane approaches to solving MIPs. At every node of a branch and cut tree, one has an option of branching on a currently fractional variable or generating one or more cutting planes. The branch and cut approach has proved very successful for solving structured 0 1 MIPs [2][3] 7] 8] as well as general integer programs [6] where the integer variables are not restricted to 0 1) and is fast becoming the method of choice for solving MIPs among practitioners. The cutting planes which we use within branch and cut are the classical Gomory cuts [9] The success of Gomory ....

....(where the integer variables are not restricted to 0 1) and is fast becoming the method of choice for solving MIPs among practitioners. The cutting planes which we use within branch and cut are the classical Gomory cuts [9] The success of Gomory cuts within branch and cut was shown recently in [2][4] We also tried lift and project cuts ( 3] instead of Gomory cuts on a pilot set of problems; these run times were comparable to, but not better than, those obtained using Gomory cuts. CPLEX (version 4.0.3) is a commercial package used for solving linear programs and MIPs. CPLEX uses the ....

Balas, E., Ceria, S., Cornu'ejols, G and Natraj, N.R. (1994), Gomory cuts revisited, to appear in OR letters .

.... (1989) Laurent (1989) Nobili and Sassano (1989) Sassano (1989) Chopra and Rao (1993) Cut polytopes: Barahona and Mahjoub (1986) Barahona et al. 1988) Conforti et al. 1990 91a,b) De Sousa and Laurent (1991) Deza et al. 1992) Deza and Laurent (1992a,b) Pulleyblank and Shepherd (1993) Balas et al. 1994b) Frequency assignment: Aardal et al. 1995) General integer and mixed 0 1 structures: Wolsey (1976a) Peled (1977) Zemel (1978) Crowder et al. 1983) Padberg et al. 1985) Van Roy and Wolsey (1985,1986,1987) Goemans (1989) Nemhauser and Wolsey (1990) Knapsack problems: Balas (1975) ....

....These inequalities were maintained in a central pool, from which one could select violated inequalities for the current subproblem. The global cuts usually work well, but to use the full power of the branch and cut algorithm, one should also be able to generate inequalities that are locally valid. Balas et al. 1994a) report on good results using branch and cut with locally valid Gomory cuts. When solving large instances it becomes important to work with a formulation that is as small as possible. One important feature is therefore to be able to delete inequalities from the active formulation and store ....

E. Balas, S. Ceria, G. Cornu' ejols and Natraj (1994a) "Gomory cuts revisited", Working Paper No. 1994--16, Carnegie Mellon University, Graduate School of Industrial Administration, Pittsburg.

....(1986) page 357. Theorem 1 Gomory (1963) There exists an implementation of Gomory s cutting plane algorithm such that after a finite number of iterations either an optimal integer solution is found, or it is proved that S = A recent discussion on Gomory cutting planes can be found in Balas et al. 1994) who incorporate the cutting plane algorithm in a branch and bound procedure and report on computational experience. Chv atal s Rounding Procedure. Chv atal (1973) studied the more general version of ILP, where the integer vectors of S are bounded and where the entries of A and b are real ....

E. Balas, S. Ceria, G. Cornu' ejols and Natraj (1994) "Gomory cuts revisited", Working Paper No. 1994--16, Carnegie Mellon University, Graduate School of Industrial Administration, Pittsburgh.

....is to use general purpose cutting planes which do not require or exploit any a priori knowledge about the structure of the problem at hand. The Gomory mixed integer cuts (see Gomory [11] which have been successfully used in a branch and cut framework (see Balas, Ceria, Cornu ejols and Natraj [4]) or the disjunctive cutting planes (see Balas [2] revisited and implemented as the equivalent lift and project cuts (see Balas, Ceria, Cornu ejols [3] fall into this category. Another possibility is to generate special purpose cutting planes that are based on the polyhedral analysis of the ....

E. Balas, S. Ceria, G. Cornu'ejols and Natraj, "Gomory Cuts Revisited", Operations Research Letters 19 (1996), 1-9.

....using the formula in Section 3, the separation problem is trivially solved. There are, however, some computational issues that need to be addressed in order to have an algorithm that uses Gomory cuts efficiently. Most of these computational (or engineering) questions have already been treated in [2] for the case of mixed 0 1 programs. Whenever we do generate Gomory cuts, we do so in rounds, or batches. A round of cuts is a set of cuts which is generated for the same fractional solution, without resolving the linear programming relaxation, from different rows of optimal tableau. In the MIPO ....

E. Balas, S. Ceria, G. Cornu'ejols and N.R. Natraj, Gomory Cuts revisited, to appear in OR Letters, (1995).

No context found.

E. Balas, S. Ceria, G. Cornueljols and N. Natraj (1996), Gomory Cuts Revisited, Operations Research Letters, 19, pp. 1--10.

No context found.

E. Balas, S. Ceria, G. Cornuejols & N. Natraj, \Gomory cuts revisited", Oper. Res. Lett., vol. 19, pp. 1-9, 1996.

No context found.

Egon Balas, Sebastian Ceria, Gerard Cornuejols, and N.R. Natraj, `Gomory cuts revisited', OR Letters, 19, 1--10 (1996).

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC