Valid Inequalities and Separation for the Symmetric Sequential Ordering Problem

Size: px
Start display at page:

Download "Valid Inequalities and Separation for the Symmetric Sequential Ordering Problem"

Transcription

1 Valid Inequalities and Separation for the Symmetric Sequential Ordering Problem Adam N. Letchford Yanjun Li Draft, April 2014 Abstract The sequential ordering problem (SOP) is the generalisation of the asymmetric TSP in which the tour must obey given precedence relations between pairs of nodes. Many solution approaches have been devised for the SOP, but the version with symmetric costs, which we call the SSOP, has received very little attention. To fill this gap, we present a new 0-1 linear programming formulation for the SSOP, and derive several new families of strong valid inequalities for it, along with efficient separation routines. Some encouraging computational results are also presented. One remarkable result is that, in the case of the SSOP, there exists an efficient separation routine for the so-called simple source-destination inequalities. No such separation routine is known for the SOP, or indeed for the symmetric or asymmetric TSP. Keywords: travelling salesman problem, sequential ordering problem, precedence constraints, cutting planes, branch-and-cut. 1 Introduction The sequential ordering problem (SOP), sometimes called the precedenceconstrained asymmetric traveling salesman problem, is defined as follows [10]. We are given a complete directed graph G = (V, A) on n nodes, with non-negative costs c ij for each arc (i, j) A. We are also given an acyclic precedence digraph H = (V, B). The task is to find a minimum cost Hamiltonian tour, starting at ending at a given depot node, that obeys the Department of Management Science, Lancaster University, Lancaster LA1 4YW, United Kingdom. A.N.Letchford@lancaster.ac.uk Krannert School of Management, Purdue University, 403 W. State Street, West Lafayette, IN , United States. li14@purdue.edu 1

2 precedences. That is, if (i, j) B, the tour must pass through node i before passing through node j. The SOP has applications in both vehicle routing and machine scheduling [1, 6, 10, 11]. It has received considerable attention in the literature, and many integer programming formulations and algorithms have been devised for it (e.g., [1, 2, 6, 7, 10, 11, 15, 16, 18, 23, 24]). On the other hand, very little work has been conducted on the symmetric sequential ordering problem (SSOP), in which, for all pairs i, j, the costs c ij and c ji are equal. Indeed, we are aware of only three works that mention the SSOP [6, 12, 13]. In this paper, we look again at the 0-1 linear programming formulations for the SSOP given in [6, 12]. We begin by proposing a modified formulation, that is defined in terms of a mixed graph, and has two copies of the depot. This formulation is more intuitive than the standard one, and easier to handle computationally. Next, we derive several new families of strong valid linear inequalities (cutting planes), that can be used to strengthen the continuous relaxation of the formulation. After that, we present exact polynomial-time separation routines for five of the families......together with heuristic separation routines for the others. Finally, we present some encouraging computational results. The paper is structured as follows. The literature review is in Section 2. The improved integer programming formulation is in Section 3. The valid inequalities are in Section 4, and the separation routines are in Section 5. The computational results are in Section 6 and some concluding remarks are in Section 7. One particularly remarkable result, given in Subsection 5.6, is that there exists an efficient exact separation routine for a huge family of valid inequalities for the SSOP, called simple mixed source-destination inequalities. No analogous separation routine is known for the SOP, or indeed for the symmetric or asymmetric TSP (see our earlier paper [17]). Throughout the paper, we assume that B contains arcs only for precedences that cannot be deduced from transitivity. We also use the following notation. For any non-depot node i, π(i) and σ(i) denote the set of predecessors and successors of i, respectively. That is, each node of π(i) must be visited before i, and each node of σ(i) must be visited after i. For any set S of non-depot nodes, we let π(s) denote the union of π(i) over all i S, and σ(s) denote the union of σ(i) over all i S. Finally, we use the convention that edges are undirected, but arcs are directed. 2

3 2 Literature Review We now review the relevant literature. We cover the standard formulation of the SOP in Subsection 2.1 and the symmetric SOP in Subsection 2.2. In Subsection 2.3, we recall some results on a family of cutting-planes for general integer linear programs, called {0, 1 2 }-cuts, since we will need them later on (in Subsections 4.4 and 5.6). 2.1 The standard formulation of the SOP Several different integer programming formulations of the SOP have been proposed [1, 6, 10, 11, 15, 16, 18, 23, 24], along with a dynamic programming formulation [7]. For the sake of brevity, we restrict attention to the standard 0-1 linear programming formulation, which is used, e.g., in [1, 2, 6, 11, 20]. Let the depot be node 1. For each arc (i, j) A, let x a be a binary variable taking the value 1 if and only if the arc is traversed. Given a set S V, let δ + (S) denote the set of arcs with their tail in S, δ (S) denote the set of arcs with their head in S, and A(S) denote the set of arcs with both head and tail in S. We have: min a A c ax a s.t. x(δ + (i)) = 1 (i V ) x(δ (i)) = 1 (i V ) x(δ + (S)) 1 (S V \ {1}) (1) x(j : S) + x(a(s)) + x(s : i) S x a {0, 1} ((i, j) B, S V \ {1, i, j})(2) (a A). The constraints (1) are the subtour elimination (SE) inequalities, well known in the literature on the ATSP (see, e.g., [5]). The constraints (2) are the so-called precedence-forcing (PF) inequalities. They ensure that the solution does not contain any directed path from j to i passing through the nodes in any set S, and thereby enforce the precedences. Note that the SE and PF inequalities are exponential in number. It is pointed out in [1] that the associated separation problems can solved in polynomial time by, e.g., solving a series of max-flow min-cut problems. In [1], it is pointed out that one can strengthen the above formulation by adding any of the known valid inequalities for the asymmetric travelling salesman problem (ATSP). These inequalities are surveyed in [5]. In [2, 11, 20], further families of inequalities are presented, that exploit the presence of the precedences. These include strengthened PF inequalities [11], π, σ, 3

4 (π, σ) and precedence-cycle-breaking inequalities [6], strengthened 2-matching inequalities [2] and strengthened D k inequalities [20]. It is shown in [1, 6, 10] that we can delete the arc (i, j) from A under either of the following conditions: j π(i) There exists a node k V \ {i, j} such that k σ(i) and k π(j). Some other, more complicated, rules for deleting arcs are given in [11]. 2.2 The symmetric SOP Now consider the symmetric version of the SOP, in which, instead of a directed graph G = (V, A) and costs c a for each a A, one is given an undirected graph G = (V, E) and costs c e for each e E. We call this problem the SSOP. To our knowledge, this problem first appeared in the MSc dissertation [12], which was concerned with a practical application involving helicopter scheduling. In [12], a 0-1 linear programming formulation is presented, and attributed to Balas and Pulleyblank. This formulation is an extension of the standard formulation of the symmetric travelling salesman problem (STSP) [9], which we now describe. For each edge e E, let x e be a binary variable, taking the value 1 if and only if edge e traversed. For each S V, let δ(s) denote the set of edges in E with exactly one end-node in S. Then the standard formulation of the STSP is as follows: min e E c ex e (3) s.t. x(δ(i)) = 2 (i V ) (4) x(δ(s)) 2 (S V : 2 S V /2) (5) x e {0, 1} (e E). (6) The constraints (4) are called degree equations, and the constraints (5) are the undirected version of the SE inequalities. To ensure that the precedence relations are obeyed, Balas and Pulleyblank added to the above formulation three rather complicated families of inequalities, called across-arc, generalised out-arc and generalised inarc inequalities. In [6], it is shown that those three families of inequalities are dominated, respectively, by the following three families: Symmetric precedence 2-cycle breaking inequalities of the form x(δ(s)) + x(δ(t )) 6 (7) 4

5 for disjoint sets S and T such that there is an arc (i, j) B with i S and j T, and also an arc (k, l) B with l S and k T. Symmetric π inequalities of the form for all S V \ {1}. Symmetric σ inequalities of the form for all S V \ {1}. x(s \ π(s) : V \ (S π(s))) 1 (8) x(s \ σ(s) : V \ (S σ(s))) 1 (9) The only other paper we are aware of concerned with the SSOP is [13]. It presents an effective constructive heuristic and a local search procedure. 2.3 {0, 1 2 }-cuts NOTE: THIS IS UNFINISHED! Another paper that will be of relevance to us is Caprara & Fischetti [8], concerned with certain cutting planes for general integer linear programs called {0, 1 2 }-cuts. A {0, 1 2 }-cut is simply a cutting plane that can be derived by summing together some given inequalities, dividing the result by two, and then rounding down. More formally, given a system of linear inequalities Ax b in non-negative integer variables x, where x Z q, A Z p q and b Z p, a {0, 1 2 }-cut is an inequality of the form λt A x λ T b for some λ {0, 1 2 }p such that λ T b / Z. In [8], it is pointed out that many of the strong inequalities known for many combinatorial optimisation problems are {0, 1 2 }-cuts with respect to a natural formulation. In particular, they mentioned that the odd closed alternating trail (odd CAT) inequalities for the ATSP, due to Balas [3], are {0, 1 2 }-cuts. They then show that the separation problem for {0, 1 2 }-cuts is N P-hard in general, but polynomially solvable in two special cases: Proposition 1 Separation... if... or its transpose is an EPT-matrix. This result can be seen as a generalisation of a result of Padberg and Rao [22], on the blossom inequalities for matching problems, due to Edmonds [?]. 5

6 (Mention also Letchford-Reinelt-Theis [19] for blossom inequalities?) For the case in which the linear system does not satisfy this condition, Caprara and Fischetti suggest to create a weakened linear system that does. This enabled them to derive a heuristic separation algorithm for the odd CAT inequalities. In Caprara et al. (2000), it is pointed out that many facet-defining inequalities for the STSP and ATSP are {0, 1 2 }-cuts. This includes the comb inequalities for the STSP, which include the 2-matching inequalities as a special case, and the SD inequalities for the ATSP, which include the comb and odd CAT inequalities as special cases. Since then, many papers have appeared that use Proposition 1 to derive exact or heuristic polynomial-time separation routines for various inequalities. Of relevance to us are the following: In [14], a heuristic separation routine is given for the cloned odd CAT inequalities for the ATSP [3]. In [17], a separation heuristic is given for the SD inequalities for the ATSP [4]. 3 A Modified Formulation of the SSOP In this section, we present a modified 0-1 linear programming formulation for the SSOP. The motivation for the modification is given in Subsection 3.1. The modification is described in Subsection 3.2, and some simple improvements are presented in Subsection Drawbacks of the original formulation Let us now consider the formulation (3) (9) in a bit more detail. Each feasible solution to the STSP constraints (4) (6) represents a tour that can be traversed in either of two directions. Such a tour is infeasible for the SSOP if and only if there exist arcs (i, j), (k, l) B such that the tour visits the nodes i, j, k and l in one of the following four orders: i, l, k, j (or the reverse), i, l, j, k (or the reverse), i, j, l, k (or the reverse), 6

7 j, i, k, l (or the reverse). One can check that the symmetric precedence 2-cycle breaking inequalities (7) cut off infeasible tours of the first two kinds, the symmetric π inequalities (8) cut off infeasible tours of the third kind, and the symmetric σ inequalities (9) cut off infeasible tours of the fourth kind. This confirms that (3) (9) is a valid formulation of the SSOP. Unfortunately, however, no separation routines are given in [6] for any of the inequalities (7) (9). So, consider the separation problem for the inequalities (7). By enumerating over all possible pairs of arcs (i, j) and (k, l) in B, one can fairly easily reduce the associated separation problem to O( B 2 ) max-flow / min-cut problems. This can be conducted in polynomial time, but could be very time-consuming when B is large. The situation is even worse when it comes to the inequalities (8) and (9), since there is no obvious way to solve the associated separation problems in polynomial time. To get around this, one could work with weakened symmetric π inequalities of the form x(s : V \ (S {i, k})) 1 for all S V \ {1} such that there exist arcs (i, j), (k, l) B with i / S, j S, k / S and l S, and weakened symmetric σ inequalities of the form x(s \ {i, k} : V \ S) 1 for all S V \ {1} such that there exist arcs (i, j), (k, l) B with i S, j / S, k S and l / S. The formulation remains valid when these weakened inequalities are used in place of the original, and the separation problems for these weakened inequalities can also be reduced to O( B 2 ) max-flow / min-cut problems. Again, however, this could be very time-consuming in practice. These considerations led us to derive a slightly different formulation, for which separation becomes much easier. 3.2 A modified formulation We have found it helpful to modify the formulation by replacing the depot node with two nodes, one representing the start of the tour and the other representing the end. Then, each feasible SSOP solution can be viewed as a Hamiltonian path from the start node to the end node. To keep notation simple, let G = (V, E) now denote the expanded graph, let n denote V, 7

8 and then let us assume that the nodes have been renumbered so that the start node is node 1 and the end node is node n. We also assume that B now contains arcs (1, i) for all nodes i that have no predecessors, and arcs (i, n) for all nodes that have no successors. Accordingly, H = (V, B) now contains a dipath from 1 to n passing through any other given node. The SSOP can now be formulated as follows: min e E c ex e (10) s.t. x(δ(i)) = 1 (i {1, n}) (11) x(δ(i)) = 2 (i V \ {1, n}) (12) x(δ(s)) 2 (S V \ {1, n}) (13) x(δ(s)) 3 (1 S V \ {n} : (i, j) B : i / S, j S) (14) x e {0, 1} (e E). (15) To see this, note that the constraints (11) (13) ensure that the solution represents a Hamiltonian path from 1 to n, and the constraints (14) impose the precedences. Clearly, the formulation (10) (15) is simpler than the formulation (3) (9). More importantly, separation becomes much easier and faster. Specifically, the separation problem for (13) can be solved in the time taken to solve one max-flow / min-cut problem, and the separation problem for (14) can be reduced to B max-flow / min-cut problems. Full details will be given in Subsections 5.1 and Improving the modified formulation It turns out that the formulation given in the previous subsection can be easily improved, in three ways. First, we can eliminate any edge {i, j} from E such that there exists a node k V \ {i, j} that must be visited between nodes i and j. (That is, if either i π(k) and k π(j), or i σ(k) and k σ(j).) The x variables associated with these edges can be eliminated (i.e., fixed at zero). Second, we can convert some edges into arcs a priori. Indeed, if the arc (i, j) is in B, the edge {i, j} E can be traversed only in the direction from i to j. Accordingly, each such edge can be replaced with the arc (i, j). In particular, each edge {1, i} can be replaced with the arc (1, i), and each edge {i, n} can be replaced with the arc (i, n). These replacements do not make the formulation stronger, but they give us additional information, which will prove to be very useful for deriving valid inequalities later on. 8

9 We denote the mixed graph that remains after making these two improvements G = (V, E B). We let L = E B denote the set of links. (Note that, between any two nodes in G, there is at most one link.) For any S, we let δ E (S) denote the set of edges in E with exactly one end-node in S, δ + (S) denote the set of arcs in B with their tail in S, δ (S) denote the set of arcs in B with their head in S, and δ(s) denote δ E (S) δ + (S) δ (S). Using this notation, the formulation resulting from the first two improvements can be written as: min l L c lx l (16) s.t. x(δ + (1)) = 1 (17) x(δ (n)) = 1 (18) x(δ(i)) = 2 (i V \ {1, n}) (19) x(δ(s)) 2 (S V \ {1, n}) (20) x(δ(s)) 3 (1 S V \ {n} : (i, j) B : i / S, j S) (21) x l {0, 1} (l L). (22) We call the constraints (20) and (21) mixed subtour elimination (mixed SE) and mixed precedence-forcing (mixed PF) inequalities, respectively. The third improvement is to add the following inequalities: x(δ + (i)) 1 (i V \ {1, n}) (23) x(δ (i)) 1 (i V \ {1, n}). (24) We call these out-bound and in-bound inequalities. If B is large, they can improve the lower bound substantially. 4 Valid Inequalities In this section, we introduce several new (and exponentially-large) families of valid inequalities for the SSOP. 4.1 Cutset inequalities Suppose that D is a dipath from 1 to n in H. For a given S, let D(S) be the set of arcs in D that have one end-node in S. Then the following cutset inequalities are valid: x(δ(s)) D(S) (S V ). (25) These dominate the mixed SE and mixed PF inequalities... 9

10 4.2 Balancing inequalities According to Ford and Fulkerson, a mixed graph is Eulerian if and only if it is connected and balanced... This leads to the following inequalities: x(δ E (S) δ + (S)) x(δ (S)) (S V \ {1, n} : S 2). (26) We call these balancing inequalities. (A similar family of inequalities was derived by Nobert & Picard [21] for a completely different vehicle routing problem, known as the mixed Chinese postman problem.) 4.3 Mixed blossom and comb inequalities We now define some valid inequalities for the SSOP that are analogous to the blossom and comb inequalities for the STSP. Let H V \ {1, n} be the handle, let T δ E (H) be the set of teeth, and let T be odd. Then we have the mixed blossom inequality: x(e(h) B(H)) + T x e H +. 2 e T (There is no point making an arc a tooth... since one would get a stronger cut by replacing the bound x a 1 with an out-bound or in-bound inequality...) (What about mixed combs? teeth?) What condition should we impose on the 4.4 Mixed SD inequalities We now define some valid inequalities for the SSOP that are analogous to the SD inequalities for the ATSP. They turn out to generalise the mixed comb inequalities... We start first with the primitive version of the these inequalities. Let H V \ {1} be the handle, let T δ E (H) be the set of teeth, S H be the source nodes and D H be the destination nodes. Assume that the 10

11 teeth, S and D have no nodes in common, and let T + S + D be odd. Then we have: x(e(h) B(H))+ T + S + D x e +x(b(s : V \H))+x(B(V \H : D)) H +. 2 e T We call these primitive mixed SD inequalities. Lemma 1 Primitive mixed SD inequalities are valid for the SSOP. Proof. Sum together the degree equations (19) over all nodes in H, the upper bounds x e 1 for all e T, the out-bound inequalities (23) over all nodes in S, and the in-bound inequalities (24) over all nodes in D, to obtain: x(e(h) B(H))+ e T x e +x(b(s))+x(b(d))+x(b(s : V \S))+x(B(V \D : D)) 2 H + T + S + D Divide this by two and round down all coefficients to the nearest integer. Note that, when S = D =, the primitive mixed SD inequalities reduce to blossom inequalities. The general class of mixed SD inequalities is obtained by applying cloning to any or all nodes in S or D, or any end-nodes of the teeth. These take the following form... When... these are analogous to the comb inequalities for the STSP. When... they are analogous to the odd CAT inequalities for the ATSP. We say that a mixed SD inequality is simple if it is obtained from a primitive SD inequality by cloning some nodes in S and D at most once. The simple mixed SD inequalities include all mixed blossom and mixed odd CAT inequalities. 5 Separation Algorithms Surprisingly, for five of the classes of inequalities that we have defined, there are polynomial-time separation algorithms! 11

12 Let E denote... let B denote... and let G denote Mixed SE inequalities These are easy to separate exactly, as usual... one can use the Nagamochi et al. (1994) algorithm. 5.2 Mixed PF inequalities These are polynomially-separable too. Solve one max-flow / min-cut problem for each arc in B. 5.3 Cutset inequalities The complexity of separation is unclear... we can always just take a violated or near-violated mixed SE or mixed PF inequality, and then solve a maximum-weight dipath algorithm to find the best possible value of the right-hand side Balancing inequalities The balancing inequalities (26) can be separated in polynomial-time by reduction to a single max-flow / min-cut problem. See [21] Mixed blossom inequalities These can be separated as in Padberg-Rao, or as in Letchford-Reinelt-Theis. In the former case, one needs to solve O( E ) max-flow / min-cut problems on a graph with... nodes and... edges. In the latter case, one needs to solve O( V ) max-flow / min-cut problems on G itself. 12

13 5.6 Simple mixed SD inequalities Observe that every feasible solution to our SSOP formulation satisfies the following linear inequalities: x(δ(i)) = 2 (i V ) (27) x(δ + (i)) 1 (i V \ {1}) (28) x(δ (i)) 1 (i V \ {1}) (29) x e 1 (e E) (30) x e 0 (e E) (31) x a 0 (a B). (32) The remarkable thing is that the linear system (27) (32) satisfies one of the two conditions in Proposition 1. Theorem 1 The matrix given by the left-hand side coefficients of the inequalities (27) (29) is an EPT-matrix. From the above theorem it follows that one can separate in polynomial time over all of the simple mixed SD inequalities mentioned in Subsection 4.4. Perhaps say something about the details of the algorithm. One must take the tree mentioned in the proof of the theorem, label some edges and nodes odd, and then add other edges representing the x variables with positive value at x. Finally, one must compute a minimum-weight odd cut in the resulting graph. Next, we need to analyse the running time. Let L be the set of links with positive value at x. The odd cut must be computed in a graph with O( L ) nodes and edges, where the number of odd edges and odd nodes seems to be O( L ). Then, using a standard max-flow algorithm, the time taken by the separation algorithm seems to be O(( L 3 log L ). 5.7 Shrinking We could probably apply shrinking to paths of 1-edges, and to 1-arcs, before running the mixed blossom and simple mixed SD separation algorithm. Then, the exact separation algorithm for mixed blossom inequalities becomes a heuristic for comb inequalities, and the exact separation algorithm for simple mixed SD inequalities becomes a heuristic for general mixed SD inequalities. But is such shrinking valid? 13

14 6 Computational Results It would be nice to compare some different bounds. Our original formulation, formulation after edge elimination, then after out-bound and in-bound inequalities, then with various combinations of cutset, balancing, mixed blossom and mixed SD inequalities... 7 Conclusion We could probably derive additional valid inequalities from conflicts between pairs of edges / arcs. The same applies to the standard SOP! References [1] N. Ascheuer, L.F. Escudero, M. Grötschel & M. Stoer (1993) A cutting plane approach to the sequential ordering problem (with applications to job scheduling in manufacturing). SIAM J. Optim., 3, [2] N. Ascheuer, M. Jünger & G. Reinelt (2000) A branch-and-cut algorithm for the asymmetric traveling salesman problem with precedence constraints. Comput. Optim. Appl., 17, [3] E. Balas (1989) The asymmetric assignment problem and some new facets of the traveling salesman polytope. SIAM J. Discr. Math., 2, [4] E. Balas & M. Fischetti (1993) A lifting procedure for the asymmetric traveling salesman polytope and a large new class of facets. Math. Program., 58, [5] E. Balas & M. Fischetti (2007) Polyhedral theory for the asymmetric traveling salesman problem. In G. Gutin & A.P. Punnen (eds.) The Traveling Salesman Problem and Its Variations, pp Springer. [6] E. Balas, M. Fischetti & W. Pulleyblank (1995) The precedenceconstrained asymmetric traveling salesman polytope. Math. Program., 68, [7] L. Bianco, A. Mingozzi & S. Ricciardelli (1994) Exact and heuristic procedures for the travelling salesman problem with precedence constraints, based on dynamic programming. INFOR, 32,

15 [8] A. Caprara & M. Fischetti (1996) {0, 1 2 }-Chvátal-Gomory cuts. Math. Program. 74, [9] G.B. Dantzig, D.R. Fulkerson & S.M. Johnson (1954) Solution of a large-scale traveling salesman problem. Oper. Res., 2, [10] L.F. Escudero (1988) An inexact algorithm for the sequential ordering problem. Eur. J. Oper. Res., 37, [11] L. Escudero, M. Guignard & K. Malik (1994) A Lagrangian relax-andcut approach for the sequential ordering problem with precedence relationships. Ann. Oper. Res., 50, [12] M.T. Fiala Timlin (1989) Precedence Constrained Routing and Helicopter Scheduling. MSc Thesis, Department of Combinatorics and Optimization, University of Waterloo, Canada. [13] M.T. Fiala Timlin & W.R. Pulleyblank (1992) Precedence constrained routing and helicopter scheduling: heuristic design. Interfaces, 22, [14] M. Fischetti & P. Toth (1997) A polyhedral approach to the asymmetric traveling salesman problem. Management Sci., 43, [15] L. Gouveia & P. Pesneau (2006) On extended formulations for the precedence constrained asymmetric traveling salesman problem. Networks, 48, [16] H. Hernández-Pérez & J.J. Salazar-González (2009) The multicommodity one-to-one pickup-and-delivery traveling salesman problem. Eur. J. Oper. Res., 196, [17] A.N. Letchford & Y. Li (2014) On the separation of {0, 1 2 }-cuts for the asymmetric TSP. Working paper, Department of Management Science, Lancaster University, United Kingdom. [18] A.N. Letchford & J.J. Salazar-González (2014) A strong multicommodity flow formulation for the sequential ordering problem. Working paper, Department of Management Science, Lancaster University, United Kingdom. [19] A.N. Letchford, G. Reinelt & D.O. Theis (2008) Odd minimum cut-sets and b-matchings revisited. SIAM J. Discr. Math., 22,

16 [20] V. Mak & A.T. Ernst (2007) New cutting-planes for the time and/or precedence-constrained ATSP and directed VRP. Math. Meth. Oper. Res., 66, [21] Y. Nobert & J.C. Picard (1996) An optimal algorithm for the mixed Chinese postman problem. Networks, 27, [22] M.W. Padberg & M.R. Rao (1982) Odd minimum cut-sets and b- matchings. Math. Oper. Res., 7, [23] S.C. Sarin, H.D. Sherali & A. Bhootra (2005) New tighter polynomial length formulations for the asymmetric traveling salesman problem with and without precedence constraints. Oper. Res. Lett., 33, [24] H.D. Sherali & P.J. Driscoll (2002) On tightening the relaxations of Miller Tucker Zemlin formulations for asymmetric traveling salesman problems. Oper. Res., 50,

Asymmetric Traveling Salesman Problem (ATSP): Models

Asymmetric Traveling Salesman Problem (ATSP): Models Asymmetric Traveling Salesman Problem (ATSP): Models Given a DIRECTED GRAPH G = (V,A) with V = {,, n} verte set A = {(i, j) : i V, j V} arc set (complete digraph) c ij = cost associated with arc (i, j)

More information

The Traveling Salesman Problem: Inequalities and Separation

The Traveling Salesman Problem: Inequalities and Separation The Traveling Salesman Problem: Inequalities and Separation Adam N. Letchford Department of Management Science, Lancaster University http://www.lancs.ac.uk/staff/letchfoa 1. The ILP formulation of the

More information

Stronger Multi-Commodity Flow Formulations of the (Capacitated) Sequential Ordering Problem

Stronger Multi-Commodity Flow Formulations of the (Capacitated) Sequential Ordering Problem Stronger Multi-Commodity Flow Formulations of the (Capacitated) Sequential Ordering Problem Adam N. Letchford Juan-José Salazar-González May 2015 Abstract The sequential ordering problem (SOP) is the generalisation

More information

Separating Simple Domino Parity Inequalities

Separating Simple Domino Parity Inequalities Separating Simple Domino Parity Inequalities Lisa Fleischer Adam Letchford Andrea Lodi DRAFT: IPCO submission Abstract In IPCO 2002, Letchford and Lodi describe an algorithm for separating simple comb

More information

Compact Formulations of the Steiner Traveling Salesman Problem and Related Problems

Compact Formulations of the Steiner Traveling Salesman Problem and Related Problems Compact Formulations of the Steiner Traveling Salesman Problem and Related Problems Adam N. Letchford Saeideh D. Nasiri Dirk Oliver Theis March 2012 Abstract The Steiner Traveling Salesman Problem (STSP)

More information

POLYNOMIAL MILP FORMULATIONS

POLYNOMIAL MILP FORMULATIONS POLYNOMIAL MILP FORMULATIONS Miller-Tucker-Zemlin (J. ACM, 1960); Gavish-Graves (MIT Tech. Report 1978) Fox-Gavish-Graves (Operations Research 1980); Wong (IEEE Conference, 1980); Claus (SIAM J. on Algebraic

More information

The traveling salesman problem

The traveling salesman problem Chapter 58 The traveling salesman problem The traveling salesman problem (TSP) asks for a shortest Hamiltonian circuit in a graph. It belongs to the most seductive problems in combinatorial optimization,

More information

Travelling Salesman Problem

Travelling Salesman Problem Travelling Salesman Problem Fabio Furini November 10th, 2014 Travelling Salesman Problem 1 Outline 1 Traveling Salesman Problem Separation Travelling Salesman Problem 2 (Asymmetric) Traveling Salesman

More information

On the separation of split cuts and related inequalities

On the separation of split cuts and related inequalities Math. Program., Ser. B 94: 279 294 (2003) Digital Object Identifier (DOI) 10.1007/s10107-002-0320-3 Alberto Caprara Adam N. Letchford On the separation of split cuts and related inequalities Received:

More information

3.7 Strong valid inequalities for structured ILP problems

3.7 Strong valid inequalities for structured ILP problems 3.7 Strong valid inequalities for structured ILP problems By studying the problem structure, we can derive strong valid inequalities yielding better approximations of conv(x ) and hence tighter bounds.

More information

New Integer Programming Formulations of the Generalized Travelling Salesman Problem

New Integer Programming Formulations of the Generalized Travelling Salesman Problem American Journal of Applied Sciences 4 (11): 932-937, 2007 ISSN 1546-9239 2007 Science Publications New Integer Programming Formulations of the Generalized Travelling Salesman Problem Petrica C. Pop Department

More information

On Disjunctive Cuts for Combinatorial Optimization

On Disjunctive Cuts for Combinatorial Optimization Journal of Combinatorial Optimization, 5, 99 315, 001 c 001 Kluwer Academic Publishers. Manufactured in The Netherlands. On Disjunctive Cuts for Combinatorial Optimization ADAM N. LETCHFORD Department

More information

Revisiting the Hamiltonian p-median problem: a new formulation on directed graphs and a branch-and-cut algorithm

Revisiting the Hamiltonian p-median problem: a new formulation on directed graphs and a branch-and-cut algorithm Revisiting the Hamiltonian p-median problem: a new formulation on directed graphs and a branch-and-cut algorithm Tolga Bektaş 1, Luís Gouveia 2, Daniel Santos 2 1 Centre for Operational Research, Management

More information

EXACT ALGORITHMS FOR THE ATSP

EXACT ALGORITHMS FOR THE ATSP EXACT ALGORITHMS FOR THE ATSP Branch-and-Bound Algorithms: Little-Murty-Sweeney-Karel (Operations Research, ); Bellmore-Malone (Operations Research, ); Garfinkel (Operations Research, ); Smith-Srinivasan-Thompson

More information

On Matroid Parity and Matching Polytopes

On Matroid Parity and Matching Polytopes On Matroid Parity and Matching Polytopes Konstantinos Kaparis, Adam N. Letchford, Ioannis Mourtos October 017 Abstract The matroid parity problem is a natural extension of the matching problem to the matroid

More information

Solving a Production Scheduling Problem as a Time-Dependent Traveling Salesman Problem

Solving a Production Scheduling Problem as a Time-Dependent Traveling Salesman Problem Solving a Production Scheduling Problem as a Time-Dependent Traveling Salesman Problem GABRIELLA STECCO Department of Applied Mathematics, University Ca Foscari of Venice, Dorsoduro n. 3825/E, 30123 Venice,

More information

Stronger Multi-Commodity Flow Formulations of the Capacitated Vehicle Routing Problem

Stronger Multi-Commodity Flow Formulations of the Capacitated Vehicle Routing Problem Stronger Multi-Commodity Flow Formulations of the Capacitated Vehicle Routing Problem Adam N. Letchford Juan-José Salazar-González June 2014 Abstract The Capacitated Vehicle Routing Problem is a much-studied

More information

Polyhedral Results for Some Constrained Arc-Routing Problems

Polyhedral Results for Some Constrained Arc-Routing Problems Adam Nicholas Letchford BA(Hons.), MSc. Polyhedral Results for Some Constrained Arc-Routing Problems PhD Thesis, December 1996. Department of Management Science, The Management School, Lancaster University.

More information

Discrete (and Continuous) Optimization WI4 131

Discrete (and Continuous) Optimization WI4 131 Discrete (and Continuous) Optimization WI4 131 Kees Roos Technische Universiteit Delft Faculteit Electrotechniek, Wiskunde en Informatica Afdeling Informatie, Systemen en Algoritmiek e-mail: C.Roos@ewi.tudelft.nl

More information

A Time Bucket Formulation for the TSP with Time Windows

A Time Bucket Formulation for the TSP with Time Windows A Time Bucket Formulation for the TSP with Time Windows Sanjeeb Dash, Oktay Günlük IBM Research Andrea Lodi, Andrea Tramontani University of Bologna November 10, 2009 Abstract The Traveling Salesman Problem

More information

A maritime version of the Travelling Salesman Problem

A maritime version of the Travelling Salesman Problem A maritime version of the Travelling Salesman Problem Enrico Malaguti, Silvano Martello, Alberto Santini May 31, 2015 Plan 1 The Capacitated TSP with Pickup and Delivery 2 The TSPPD with Draught Limits

More information

Introduction to Mathematical Programming IE406. Lecture 21. Dr. Ted Ralphs

Introduction to Mathematical Programming IE406. Lecture 21. Dr. Ted Ralphs Introduction to Mathematical Programming IE406 Lecture 21 Dr. Ted Ralphs IE406 Lecture 21 1 Reading for This Lecture Bertsimas Sections 10.2, 10.3, 11.1, 11.2 IE406 Lecture 21 2 Branch and Bound Branch

More information

Facets from Gadgets. Adam N. Letchford Anh Vu. December 2017

Facets from Gadgets. Adam N. Letchford Anh Vu. December 2017 Facets from Gadgets Adam N. Letchford Anh Vu December 2017 Abstract We present a new tool for generating cutting planes for N P-hard combinatorial optimisation problems. It is based on the concept of gadgets

More information

Solving the Asymmetric Travelling Salesman Problem with time windows by branch-and-cut

Solving the Asymmetric Travelling Salesman Problem with time windows by branch-and-cut Math. Program., Ser. A 90: 475 506 (2001) Digital Object Identifier (DOI) 10.1007/s101070100218 Norbert Ascheuer Matteo Fischetti Martin Grötschel Solving the Asymmetric Travelling Salesman Problem with

More information

The Multiple Traveling Salesman Problem with Time Windows: Bounds for the Minimum Number of Vehicles

The Multiple Traveling Salesman Problem with Time Windows: Bounds for the Minimum Number of Vehicles The Multiple Traveling Salesman Problem with Time Windows: Bounds for the Minimum Number of Vehicles Snežana Mitrović-Minić Ramesh Krishnamurti School of Computing Science, Simon Fraser University, Burnaby,

More information

3.7 Cutting plane methods

3.7 Cutting plane methods 3.7 Cutting plane methods Generic ILP problem min{ c t x : x X = {x Z n + : Ax b} } with m n matrix A and n 1 vector b of rationals. According to Meyer s theorem: There exists an ideal formulation: conv(x

More information

Solving Elementary Shortest-Path Problems as Mixed-Integer Programs

Solving Elementary Shortest-Path Problems as Mixed-Integer Programs Gutenberg School of Management and Economics Discussion Paper Series Solving Elementary Shortest-Path Problems as Mixed-Integer Programs Michael Drexl and Stefan Irnich Januar 2012 Discussion paper number

More information

A Note on Representations of Linear Inequalities in Non-Convex Mixed-Integer Quadratic Programs

A Note on Representations of Linear Inequalities in Non-Convex Mixed-Integer Quadratic Programs A Note on Representations of Linear Inequalities in Non-Convex Mixed-Integer Quadratic Programs Adam N. Letchford Daniel J. Grainger To appear in Operations Research Letters Abstract In the literature

More information

Applied Integer Programming: Modeling and Solution

Applied Integer Programming: Modeling and Solution Applied Integer Programming: Modeling and Solution Chen, Batson, Dang Section 6. - 6.3 Blekinge Institute of Technology April 5, 05 Modeling Combinatorical Optimization Problems II Traveling Salesman Problem

More information

A comparative analysis of several asymmetric traveling salesman problem formulations

A comparative analysis of several asymmetric traveling salesman problem formulations Computers & Operations Research 36 (2009) 637 654 www.elsevier.com/locate/cor Invited review A comparative analysis of several asymmetric traveling salesman problem formulations Temel Öncan a, İ. Kuban

More information

A Branch-and-Cut Algorithm for the Dial-a-Ride Problem

A Branch-and-Cut Algorithm for the Dial-a-Ride Problem A Branch-and-Cut Algorithm for the Dial-a-Ride Problem JEAN-FRANÇOIS CORDEAU Canada Research Chair in Distribution Management, HEC Montréal 3000, chemin de la Côte-Sainte-Catherine Montréal, Canada H3T

More information

Bounds on the Traveling Salesman Problem

Bounds on the Traveling Salesman Problem Bounds on the Traveling Salesman Problem Sean Zachary Roberson Texas A&M University MATH 613, Graph Theory A common routing problem is as follows: given a collection of stops (for example, towns, stations,

More information

3.8 Strong valid inequalities

3.8 Strong valid inequalities 3.8 Strong valid inequalities By studying the problem structure, we can derive strong valid inequalities which lead to better approximations of the ideal formulation conv(x ) and hence to tighter bounds.

More information

The Traveling Salesman Problem with Pickup and Delivery. A polyhedral approach. IBM Research - Australia. Irina Dumitrescu

The Traveling Salesman Problem with Pickup and Delivery. A polyhedral approach. IBM Research - Australia. Irina Dumitrescu Australia The Traveling Salesman Problem with Pickup and Delivery A polyhedral approach Irina Dumitrescu Jean-Francois Cordeau, Gilbert Laporte, Stefan Ropke The TSP with Pickup and Delivery (TSPPD) Given:

More information

A simple LP relaxation for the Asymmetric Traveling Salesman Problem

A simple LP relaxation for the Asymmetric Traveling Salesman Problem A simple LP relaxation for the Asymmetric Traveling Salesman Problem Thành Nguyen Cornell University, Center for Applies Mathematics 657 Rhodes Hall, Ithaca, NY, 14853,USA thanh@cs.cornell.edu Abstract.

More information

Modeling with Integer Programming

Modeling with Integer Programming Modeling with Integer Programg Laura Galli December 18, 2014 We can use 0-1 (binary) variables for a variety of purposes, such as: Modeling yes/no decisions Enforcing disjunctions Enforcing logical conditions

More information

A new family of facet defining inequalities for the maximum edge-weighted clique problem

A new family of facet defining inequalities for the maximum edge-weighted clique problem A new family of facet defining inequalities for the maximum edge-weighted clique problem Franklin Djeumou Fomeni June 2016 Abstract This paper considers a family of cutting planes, recently developed for

More information

Teaching Integer Programming Formulations Using the Traveling Salesman Problem

Teaching Integer Programming Formulations Using the Traveling Salesman Problem SIAM REVIEW Vol. 45, No. 1, pp. 116 123 c 2003 Society for Industrial and Applied Mathematics Teaching Integer Programming Formulations Using the Traveling Salesman Problem Gábor Pataki Abstract. We designed

More information

Time Dependent Traveling Salesman Problem with Time Windows: Properties and an Exact Algorithm

Time Dependent Traveling Salesman Problem with Time Windows: Properties and an Exact Algorithm Time Dependent Traveling Salesman Problem with Time Windows: Properties and an Exact Algorithm Anna Arigliano, Gianpaolo Ghiani, Antonio Grieco, Emanuela Guerriero Dipartimento di Ingegneria dell Innovazione,

More information

Deciding Emptiness of the Gomory-Chvátal Closure is NP-Complete, Even for a Rational Polyhedron Containing No Integer Point

Deciding Emptiness of the Gomory-Chvátal Closure is NP-Complete, Even for a Rational Polyhedron Containing No Integer Point Deciding Emptiness of the Gomory-Chvátal Closure is NP-Complete, Even for a Rational Polyhedron Containing No Integer Point Gérard Cornuéjols 1 and Yanjun Li 2 1 Tepper School of Business, Carnegie Mellon

More information

A Dynamic Programming Heuristic for the Quadratic Knapsack Problem

A Dynamic Programming Heuristic for the Quadratic Knapsack Problem A Dynamic Programming Heuristic for the Quadratic Knapsack Problem Franklin Djeumou Fomeni Adam N. Letchford March 2012 Abstract It is well known that the standard (linear) knapsack problem can be solved

More information

A Polyhedral Study of the Asymmetric Traveling Salesman Problem with Time Windows

A Polyhedral Study of the Asymmetric Traveling Salesman Problem with Time Windows A Polyhedral Study of the Asymmetric Traveling Salesman Problem with Time Windows Norbert Ascheuer Intranetz GmbH, Bergstr. 22, 10115 Berlin, Germany Matteo Fischetti Dipartimento di Elettronica ed Informatica,

More information

La petite et la grande histoire du problème du voyageur de commerce

La petite et la grande histoire du problème du voyageur de commerce La petite et la grande histoire du problème du voyageur de commerce par Gilbert Laporte Chaire de recherche du Canada en distributique, Centre de recherche sur les transports (CRT) et GERAD HEC Montréal,

More information

Models and algorithms for the Asymmetric Traveling Salesman Problem: an experimental comparison

Models and algorithms for the Asymmetric Traveling Salesman Problem: an experimental comparison Downloaded from orbit.dtu.dk on: Nov 03, 2018 Models and algorithms for the Asymmetric Traveling Salesman Problem: an experimental comparison Roberti, Roberto; Toth, Paolo Published in: EURO Journal on

More information

Cutting Planes for RLT Relaxations of Mixed 0-1 Polynomial Programs

Cutting Planes for RLT Relaxations of Mixed 0-1 Polynomial Programs Cutting Planes for RLT Relaxations of Mixed 0-1 Polynomial Programs Franklin Djeumou Fomeni Konstantinos Kaparis Adam N. Letchford December 2013 Abstract The Reformulation-Linearization Technique (RLT),

More information

Combinatorial optimization problems

Combinatorial optimization problems Combinatorial optimization problems Heuristic Algorithms Giovanni Righini University of Milan Department of Computer Science (Crema) Optimization In general an optimization problem can be formulated as:

More information

Memorandum COSOR 95-19, 1995, Eindhoven University of Technology

Memorandum COSOR 95-19, 1995, Eindhoven University of Technology Memorandum COSOR 95-19, 1995, Eindhoven University of Technology A polyhedral approach to the delivery man problem C.A. van Eijl Abstract We propose a mixed integer programming formulation for the delivery

More information

Formulations and Valid Inequalities for the Heterogeneous Vehicle Routing Problem

Formulations and Valid Inequalities for the Heterogeneous Vehicle Routing Problem Math. Program., Ser. A 106, 365 390 (2006) Digital Object Identifier (DOI) 10.1007/s10107-005-0611-6 Hande Yaman Formulations and Valid Inequalities for the Heterogeneous Vehicle Routing Problem Received:

More information

Chapter 3: Discrete Optimization Integer Programming

Chapter 3: Discrete Optimization Integer Programming Chapter 3: Discrete Optimization Integer Programming Edoardo Amaldi DEIB Politecnico di Milano edoardo.amaldi@polimi.it Website: http://home.deib.polimi.it/amaldi/opt-16-17.shtml Academic year 2016-17

More information

2 Notation and Preliminaries

2 Notation and Preliminaries On Asymmetric TSP: Transformation to Symmetric TSP and Performance Bound Ratnesh Kumar Haomin Li epartment of Electrical Engineering University of Kentucky Lexington, KY 40506-0046 Abstract We show that

More information

Reverse Multistar Inequalities and Vehicle Routing Problems with lower bound capacities. Luis Gouveia, Jorge Riera and Juan-José Salazar-González

Reverse Multistar Inequalities and Vehicle Routing Problems with lower bound capacities. Luis Gouveia, Jorge Riera and Juan-José Salazar-González Reverse Multistar Inequalities and Vehicle Routing Problems with lower bound capacities Luis Gouveia, Jorge Riera and Juan-José Salazar-González CIO Working Paper 13/2009 Reverse Multistar Inequalities

More information

ON COST MATRICES WITH TWO AND THREE DISTINCT VALUES OF HAMILTONIAN PATHS AND CYCLES

ON COST MATRICES WITH TWO AND THREE DISTINCT VALUES OF HAMILTONIAN PATHS AND CYCLES ON COST MATRICES WITH TWO AND THREE DISTINCT VALUES OF HAMILTONIAN PATHS AND CYCLES SANTOSH N. KABADI AND ABRAHAM P. PUNNEN Abstract. Polynomially testable characterization of cost matrices associated

More information

MVE165/MMG631 Linear and integer optimization with applications Lecture 8 Discrete optimization: theory and algorithms

MVE165/MMG631 Linear and integer optimization with applications Lecture 8 Discrete optimization: theory and algorithms MVE165/MMG631 Linear and integer optimization with applications Lecture 8 Discrete optimization: theory and algorithms Ann-Brith Strömberg 2017 04 07 Lecture 8 Linear and integer optimization with applications

More information

Week 8. 1 LP is easy: the Ellipsoid Method

Week 8. 1 LP is easy: the Ellipsoid Method Week 8 1 LP is easy: the Ellipsoid Method In 1979 Khachyan proved that LP is solvable in polynomial time by a method of shrinking ellipsoids. The running time is polynomial in the number of variables n,

More information

Chapter 3: Discrete Optimization Integer Programming

Chapter 3: Discrete Optimization Integer Programming Chapter 3: Discrete Optimization Integer Programming Edoardo Amaldi DEIB Politecnico di Milano edoardo.amaldi@polimi.it Sito web: http://home.deib.polimi.it/amaldi/ott-13-14.shtml A.A. 2013-14 Edoardo

More information

The Traveling Salesman Problem: An Overview. David P. Williamson, Cornell University Ebay Research January 21, 2014

The Traveling Salesman Problem: An Overview. David P. Williamson, Cornell University Ebay Research January 21, 2014 The Traveling Salesman Problem: An Overview David P. Williamson, Cornell University Ebay Research January 21, 2014 (Cook 2012) A highly readable introduction Some terminology (imprecise) Problem Traditional

More information

Integer Programming ISE 418. Lecture 8. Dr. Ted Ralphs

Integer Programming ISE 418. Lecture 8. Dr. Ted Ralphs Integer Programming ISE 418 Lecture 8 Dr. Ted Ralphs ISE 418 Lecture 8 1 Reading for This Lecture Wolsey Chapter 2 Nemhauser and Wolsey Sections II.3.1, II.3.6, II.4.1, II.4.2, II.5.4 Duality for Mixed-Integer

More information

arxiv: v1 [cs.cc] 5 Dec 2018

arxiv: v1 [cs.cc] 5 Dec 2018 Consistency for 0 1 Programming Danial Davarnia 1 and J. N. Hooker 2 1 Iowa state University davarnia@iastate.edu 2 Carnegie Mellon University jh38@andrew.cmu.edu arxiv:1812.02215v1 [cs.cc] 5 Dec 2018

More information

Formulations and Algorithms for Minimum Connected Dominating Set Problems

Formulations and Algorithms for Minimum Connected Dominating Set Problems Formulations and Algorithms for Minimum Connected Dominating Set Problems Abilio Lucena 1 Alexandre Salles da Cunha 2 Luidi G. Simonetti 3 1 Universidade Federal do Rio de Janeiro 2 Universidade Federal

More information

Decision Diagrams for Sequencing and Scheduling

Decision Diagrams for Sequencing and Scheduling Decision Diagrams for Sequencing and Scheduling Willem-Jan van Hoeve Tepper School of Business Carnegie Mellon University www.andrew.cmu.edu/user/vanhoeve/mdd/ Plan What can MDDs do for Combinatorial Optimization?

More information

Cutting Planes for RLT Relaxations of Mixed 0-1 Polynomial Programs

Cutting Planes for RLT Relaxations of Mixed 0-1 Polynomial Programs Cutting Planes for RLT Relaxations of Mixed 0-1 Polynomial Programs Franklin Djeumou Fomeni Konstantinos Kaparis Adam N. Letchford To Appear in Mathematical Programming Abstract The Reformulation-Linearization

More information

Decomposition-based Methods for Large-scale Discrete Optimization p.1

Decomposition-based Methods for Large-scale Discrete Optimization p.1 Decomposition-based Methods for Large-scale Discrete Optimization Matthew V Galati Ted K Ralphs Department of Industrial and Systems Engineering Lehigh University, Bethlehem, PA, USA Départment de Mathématiques

More information

CHAPTER 3 FUNDAMENTALS OF COMPUTATIONAL COMPLEXITY. E. Amaldi Foundations of Operations Research Politecnico di Milano 1

CHAPTER 3 FUNDAMENTALS OF COMPUTATIONAL COMPLEXITY. E. Amaldi Foundations of Operations Research Politecnico di Milano 1 CHAPTER 3 FUNDAMENTALS OF COMPUTATIONAL COMPLEXITY E. Amaldi Foundations of Operations Research Politecnico di Milano 1 Goal: Evaluate the computational requirements (this course s focus: time) to solve

More information

An Exact Algorithm for the Traveling Salesman Problem with Deliveries and Collections

An Exact Algorithm for the Traveling Salesman Problem with Deliveries and Collections An Exact Algorithm for the Traveling Salesman Problem with Deliveries and Collections R. Baldacci DISMI, University of Modena and Reggio E., V.le Allegri 15, 42100 Reggio E., Italy E. Hadjiconstantinou

More information

to work with) can be solved by solving their LP relaxations with the Simplex method I Cutting plane algorithms, e.g., Gomory s fractional cutting

to work with) can be solved by solving their LP relaxations with the Simplex method I Cutting plane algorithms, e.g., Gomory s fractional cutting Summary so far z =max{c T x : Ax apple b, x 2 Z n +} I Modeling with IP (and MIP, and BIP) problems I Formulation for a discrete set that is a feasible region of an IP I Alternative formulations for the

More information

Polyhedral Approach to Integer Linear Programming. Tepper School of Business Carnegie Mellon University, Pittsburgh

Polyhedral Approach to Integer Linear Programming. Tepper School of Business Carnegie Mellon University, Pittsburgh Polyhedral Approach to Integer Linear Programming Gérard Cornuéjols Tepper School of Business Carnegie Mellon University, Pittsburgh 1 / 30 Brief history First Algorithms Polynomial Algorithms Solving

More information

Integer Linear Programming (ILP)

Integer Linear Programming (ILP) Integer Linear Programming (ILP) Zdeněk Hanzálek, Přemysl Šůcha hanzalek@fel.cvut.cz CTU in Prague March 8, 2017 Z. Hanzálek (CTU) Integer Linear Programming (ILP) March 8, 2017 1 / 43 Table of contents

More information

Discrete Optimization 2010 Lecture 7 Introduction to Integer Programming

Discrete Optimization 2010 Lecture 7 Introduction to Integer Programming Discrete Optimization 2010 Lecture 7 Introduction to Integer Programming Marc Uetz University of Twente m.uetz@utwente.nl Lecture 8: sheet 1 / 32 Marc Uetz Discrete Optimization Outline 1 Intro: The Matching

More information

Exact and Heuristic Algorithms for the Symmetric and Asymmetric Vehicle Routing Problem with Backhauls

Exact and Heuristic Algorithms for the Symmetric and Asymmetric Vehicle Routing Problem with Backhauls Exact and Heuristic Algorithms for the Symmetric and Asymmetric Vehicle Routing Problem with Backhauls Paolo Toth, Daniele Vigo ECCO IX - Dublin 1996 Exact and Heuristic Algorithms for VRPB 1 Vehicle Routing

More information

On the matrix-cut rank of polyhedra

On the matrix-cut rank of polyhedra On the matrix-cut rank of polyhedra William Cook and Sanjeeb Dash Computational and Applied Mathematics Rice University August 5, 00 Abstract Lovász and Schrijver (99) described a semi-definite operator

More information

Branch-and-cut Approaches for Chance-constrained Formulations of Reliable Network Design Problems

Branch-and-cut Approaches for Chance-constrained Formulations of Reliable Network Design Problems Branch-and-cut Approaches for Chance-constrained Formulations of Reliable Network Design Problems Yongjia Song James R. Luedtke August 9, 2012 Abstract We study solution approaches for the design of reliably

More information

The Maximum Flow Problem with Disjunctive Constraints

The Maximum Flow Problem with Disjunctive Constraints The Maximum Flow Problem with Disjunctive Constraints Ulrich Pferschy Joachim Schauer Abstract We study the maximum flow problem subject to binary disjunctive constraints in a directed graph: A negative

More information

Low-Complexity Algorithms for Sequencing Jobs with a Fixed Number of Job-Classes

Low-Complexity Algorithms for Sequencing Jobs with a Fixed Number of Job-Classes Low-Complexity Algorithms for Sequencing Jobs with a Fixed Number of Job-Classes Jack A.A. van der Veen Nijenrode University - The Netherlands School of Business Straatweg 25, 3621 BG Breukelen The Netherlands

More information

On the Integrality Ratio for the Asymmetric Traveling Salesman Problem

On the Integrality Ratio for the Asymmetric Traveling Salesman Problem On the Integrality Ratio for the Asymmetric Traveling Salesman Problem Moses Charikar Dept. of Computer Science, Princeton University, 35 Olden St., Princeton, NJ 08540 email: moses@cs.princeton.edu Michel

More information

Part III: Traveling salesman problems

Part III: Traveling salesman problems Transportation Logistics Part III: Traveling salesman problems c R.F. Hartl, S.N. Parragh 1/74 Motivation Motivation Why do we study the TSP? it easy to formulate it is a difficult problem many significant

More information

When the Gomory-Chvátal Closure Coincides with the Integer Hull

When the Gomory-Chvátal Closure Coincides with the Integer Hull Date for Revised Manuscript: December 19, 2017 When the Gomory-Chvátal Closure Coincides with the Integer Hull Gérard Cornuéjols Yanjun Li Abstract Gomory-Chvátal cuts are prominent in integer programming.

More information

A Mixed-Integer Linear Program for the Traveling Salesman Problem with Structured Time Windows

A Mixed-Integer Linear Program for the Traveling Salesman Problem with Structured Time Windows A Mixed-Integer Linear Program for the Traveling Salesman Problem with Structured Time Windows Philipp Hungerländer Christian Truden 5th January 2017 Abstract In this extended abstract we introduce the

More information

Algorithms and Complexity theory

Algorithms and Complexity theory Algorithms and Complexity theory Thibaut Barthelemy Some slides kindly provided by Fabien Tricoire University of Vienna WS 2014 Outline 1 Algorithms Overview How to write an algorithm 2 Complexity theory

More information

Week Cuts, Branch & Bound, and Lagrangean Relaxation

Week Cuts, Branch & Bound, and Lagrangean Relaxation Week 11 1 Integer Linear Programming This week we will discuss solution methods for solving integer linear programming problems. I will skip the part on complexity theory, Section 11.8, although this is

More information

Not Every GTSP Facet Induces an STSP Facet

Not Every GTSP Facet Induces an STSP Facet Not Every GTSP Facet Induces an STSP Facet Marcus Oswald, Gerhard Reinelt, and Dirk Oliver Theis Institute of Computer Science University of Heidelberg Im Neuenheimer Feld 38, 9120 Heidelberg, Germany

More information

Cutting Planes in SCIP

Cutting Planes in SCIP Cutting Planes in SCIP Kati Wolter Zuse-Institute Berlin Department Optimization Berlin, 6th June 2007 Outline 1 Cutting Planes in SCIP 2 Cutting Planes for the 0-1 Knapsack Problem 2.1 Cover Cuts 2.2

More information

TSP Cuts Which Do Not Conform to the Template Paradigm

TSP Cuts Which Do Not Conform to the Template Paradigm TSP Cuts Which Do Not Conform to the Template Paradigm David Applegate 1, Robert Bixby 2, Vašek Chvátal 3, and William Cook 4 1 Algorithms and Optimization Department, AT&T Labs Research, Florham Park,

More information

Part III: Traveling salesman problems

Part III: Traveling salesman problems Transportation Logistics Part III: Traveling salesman problems c R.F. Hartl, S.N. Parragh 1/282 Motivation Motivation Why do we study the TSP? c R.F. Hartl, S.N. Parragh 2/282 Motivation Motivation Why

More information

Decomposition and Reformulation in Integer Programming

Decomposition and Reformulation in Integer Programming and Reformulation in Integer Programming Laurence A. WOLSEY 7/1/2008 / Aussois and Reformulation in Integer Programming Outline 1 Resource 2 and Reformulation in Integer Programming Outline Resource 1

More information

Outline. Outline. Outline DMP204 SCHEDULING, TIMETABLING AND ROUTING. 1. Scheduling CPM/PERT Resource Constrained Project Scheduling Model

Outline. Outline. Outline DMP204 SCHEDULING, TIMETABLING AND ROUTING. 1. Scheduling CPM/PERT Resource Constrained Project Scheduling Model Outline DMP204 SCHEDULING, TIMETABLING AND ROUTING Lecture 3 and Mixed Integer Programg Marco Chiarandini 1. Resource Constrained Project Model 2. Mathematical Programg 2 Outline Outline 1. Resource Constrained

More information

On the Polyhedral Structure of a Multi Item Production Planning Model with Setup Times

On the Polyhedral Structure of a Multi Item Production Planning Model with Setup Times CORE DISCUSSION PAPER 2000/52 On the Polyhedral Structure of a Multi Item Production Planning Model with Setup Times Andrew J. Miller 1, George L. Nemhauser 2, and Martin W.P. Savelsbergh 2 November 2000

More information

Resource Constrained Project Scheduling Linear and Integer Programming (1)

Resource Constrained Project Scheduling Linear and Integer Programming (1) DM204, 2010 SCHEDULING, TIMETABLING AND ROUTING Lecture 3 Resource Constrained Project Linear and Integer Programming (1) Marco Chiarandini Department of Mathematics & Computer Science University of Southern

More information

On the Lovász Theta Function and Some Variants

On the Lovász Theta Function and Some Variants On the Lovász Theta Function and Some Variants Laura Galli Adam N. Letchford March 2017. To appear in Discrete Optimization. Abstract The Lovász theta function of a graph is a well-known upper bound on

More information

Improved Approximation Algorithms for Metric Maximum ATSP and Maximum 3-Cycle Cover Problems

Improved Approximation Algorithms for Metric Maximum ATSP and Maximum 3-Cycle Cover Problems Improved Approximation Algorithms for Metric Maximum ATSP and Maximum 3-Cycle Cover Problems Markus Bläser 1 Computer Science, Saarland University L. Shankar Ram Institut für Theoretische Informatik, ETH

More information

Lecture 8: Column Generation

Lecture 8: Column Generation Lecture 8: Column Generation (3 units) Outline Cutting stock problem Classical IP formulation Set covering formulation Column generation A dual perspective Vehicle routing problem 1 / 33 Cutting stock

More information

The Probabilistic Pickup and Delivery Problem

The Probabilistic Pickup and Delivery Problem The Probabilistic Pickup and Delivery Problem Enrique Benavent 1, M. Landete 2, J.J. Salazar 3, G. Tirado 4 1 Departament d'estadística i Investigació Operativa, Universitat de València, Spain 2 Universidad

More information

Projection Results for the k-partition Problem

Projection Results for the k-partition Problem Projection Results for the k-partition Problem Jamie Fairbrother Adam N. Letchford February 2016 Abstract The k-partition problem is an N P-hard combinatorial optimisation problem with many applications.

More information

Partial Path Column Generation for the Vehicle Routing Problem with Time Windows

Partial Path Column Generation for the Vehicle Routing Problem with Time Windows Partial Path Column Generation for the Vehicle Routing Problem with Time Windows Bjørn Petersen & Mads Kehlet Jepsen } DIKU Department of Computer Science, University of Copenhagen Universitetsparken 1,

More information

The Cutting Plane Method is Polynomial for Perfect Matchings

The Cutting Plane Method is Polynomial for Perfect Matchings The Cutting Plane Method is Polynomial for Perfect Matchings Karthekeyan Chandrasekaran 1, László A. Végh 2, and Santosh S. Vempala 1 1 College of Computing, Georgia Institute of Technology 2 Department

More information

Lower Bounds on the Sizes of Integer Programs Without Additional Variables

Lower Bounds on the Sizes of Integer Programs Without Additional Variables Lower Bounds on the Sizes of Integer Programs Without Additional Variables Volker Kaibel and Stefan Weltge Otto-von-Guericke-Universität Magdeburg, Germany {kaibel,weltge}@ovgu.de Abstract. For a given

More information

Compact vs. exponential-size LP relaxations

Compact vs. exponential-size LP relaxations Operations Research Letters 30 (2002) 57 65 Operations Research Letters www.elsevier.com/locate/dsw Compact vs. exponential-size LP relaxations Robert D. Carr a;, Giuseppe Lancia b;c a Sandia National

More information

ON THE INTEGRALITY OF THE UNCAPACITATED FACILITY LOCATION POLYTOPE. 1. Introduction

ON THE INTEGRALITY OF THE UNCAPACITATED FACILITY LOCATION POLYTOPE. 1. Introduction ON THE INTEGRALITY OF THE UNCAPACITATED FACILITY LOCATION POLYTOPE MOURAD BAÏOU AND FRANCISCO BARAHONA Abstract We study a system of linear inequalities associated with the uncapacitated facility location

More information

18 hours nodes, first feasible 3.7% gap Time: 92 days!! LP relaxation at root node: Branch and bound

18 hours nodes, first feasible 3.7% gap Time: 92 days!! LP relaxation at root node: Branch and bound The MIP Landscape 1 Example 1: LP still can be HARD SGM: Schedule Generation Model Example 157323 1: LP rows, still can 182812 be HARD columns, 6348437 nzs LP relaxation at root node: 18 hours Branch and

More information

Matteo Fischetti, DEI, University of Padova. COFIN Matheon workshop, Villa Vigoni (Como), May 2006

Matteo Fischetti, DEI, University of Padova. COFIN Matheon workshop, Villa Vigoni (Como), May 2006 Matteo Fischetti, DEI, University of Padova COFIN Matheon workshop, Villa Vigoni (Como), May 2006 1 MIP solvers for hard optimization problems Mixed-integer linear programming (MIP) plays a central role

More information

Cut-First Branch-and-Price-Second for the CARP Workshop on Large Scale Optimization 2012 Vevey, Switzerland

Cut-First Branch-and-Price-Second for the CARP Workshop on Large Scale Optimization 2012 Vevey, Switzerland Cut-First Branch-and-Price-Second for the CARP Workshop on Large Scale Optimization 2012 Vevey, Switzerland Claudia Bode and Stefan Irnich {claudia.bode,irnich}@uni-mainz.de Chair for Logistics Management

More information