Scippy

SCIP

Solving Constraint Integer Programs

ConshdlrSubtour.cpp
Go to the documentation of this file.
1/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */
2/* */
3/* This file is part of the program and library */
4/* SCIP --- Solving Constraint Integer Programs */
5/* */
6/* Copyright (c) 2002-2024 Zuse Institute Berlin (ZIB) */
7/* */
8/* Licensed under the Apache License, Version 2.0 (the "License"); */
9/* you may not use this file except in compliance with the License. */
10/* You may obtain a copy of the License at */
11/* */
12/* http://www.apache.org/licenses/LICENSE-2.0 */
13/* */
14/* Unless required by applicable law or agreed to in writing, software */
15/* distributed under the License is distributed on an "AS IS" BASIS, */
16/* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. */
17/* See the License for the specific language governing permissions and */
18/* limitations under the License. */
19/* */
20/* You should have received a copy of the Apache-2.0 license */
21/* along with SCIP; see the file LICENSE. If not visit scipopt.org. */
22/* */
23/* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * */
24
25/**@file ConshdlrSubtour.cpp
26 * @brief Subtour elimination constraint handler for TSP problems, written in C++
27 * @author Timo Berthold
28 */
29
30/*---+----1----+----2----+----3----+----4----+----5----+----6----+----7----+----8----+----9----+----0----+----1----+----2*/
31
32#include <cassert>
33#include <string>
34#include <iostream>
35#include "ConshdlrSubtour.h"
36#include "GomoryHuTree.h"
37
38#include "objscip/objscip.h"
39
40#include "scip/cons_linear.h"
41
42#include "GomoryHuTree.h"
43
44using namespace tsp;
45using namespace scip;
46using namespace std;
47
48/** data structure for subtour elimination constraints */
49struct SCIP_ConsData
50{
51 GRAPH* graph;
52};
53
54
55/** checks whether proposed solution contains a subtour
56 *
57 * We assume that the solution is binary.
58 */
59static
61 SCIP* scip, /**< SCIP data structure */
62 GRAPH* graph, /**< underlying graph */
63 SCIP_SOL* sol /**< proposed solution */
64 )
65{
66 GRAPHNODE* node;
67 GRAPHNODE* startnode;
68 GRAPHEDGE* edge;
69 GRAPHEDGE* nextedge;
70 GRAPHEDGE* lastedge = NULL;
71 int tourlength = 0;
72
73 assert(scip != NULL);
74 assert(graph != NULL);
75
76 if( graph->nnodes <= 1 )
77 return FALSE;
78
79 startnode = &graph->nodes[0];
80 node = startnode;
81
82 /* follow the (sub?)tour until you come back to the startnode */
83 do
84 {
85 edge = node->first_edge;
86 nextedge = NULL;
87
88 /* look for an outgoing edge to proceed */
89 while( edge != NULL )
90 {
91 /* if a new edge with value 1 is found, we proceed */
92 if( edge->back != lastedge && SCIPgetSolVal(scip, sol, edge->var) > 0.5 )
93 {
94 ++tourlength;
95
96 /* if we found a subtour without the starting node, e.g. 0 - 1 - 2 - 3 - 1 - 2 - ...;
97 * this can only happen, if the degree constraints are violated; */
98 if( nextedge != NULL || tourlength > graph->nnodes )
99 return TRUE;
100
101 nextedge = edge;
102
103 /* only use the first edge for the start node */
104 if( node == startnode )
105 break;
106 }
107
108 edge = edge->next;
109 }
110
111 /* no outgoing edge found in the solution: the degree constraints must be violated; abort! */
112 if( nextedge == NULL )
113 return TRUE;
114
115 node = nextedge->adjac;
116 lastedge = nextedge;
117 }
118 while( node != startnode );
119
120 assert(tourlength <= graph->nnodes);
121
122 return ( graph->nnodes != tourlength );
123}
124
125
126/** separates subtour elemination cuts */
127static
129 SCIP* scip, /**< SCIP data structure */
130 SCIP_CONSHDLR* conshdlr, /**< the constraint handler itself */
131 SCIP_CONS** conss, /**< array of constraints to process */
132 int nconss, /**< number of constraints to process */
133 int nusefulconss, /**< number of useful (non-obsolete) constraints to process */
134 SCIP_SOL* sol, /**< primal solution that should be separated */
135 SCIP_Bool enforce, /**< whether we are in enforcing */
136 SCIP_RESULT* result /**< pointer to store the result of the separation call */
137 )
138{ /*lint --e{715}*/
139 assert(result != NULL);
140
141 *result = SCIP_DIDNOTFIND;
142
143 for(int c = 0; c < nusefulconss && *result != SCIP_CUTOFF; ++c)
144 {
145 // get all required structures
146 SCIP_CONSDATA* consdata;
147 GRAPH* graph;
148
149 consdata = SCIPconsGetData(conss[c]);
150 assert(consdata != NULL);
151
152 graph = consdata->graph;
153 assert(graph != NULL);
154
155 // store the suggested, but infeasible solution into the capacity of the edges
156 for(int i = 0; i < graph->nedges; i++)
157 {
158 double cap = SCIPgetSolVal(scip, sol, graph->edges[i].var);
159 graph->edges[i].rcap = cap;
160 graph->edges[i].cap = cap;
161 graph->edges[i].back->rcap = cap;
162 graph->edges[i].back->cap = cap;
163 }
164
165 SCIP_Bool** cuts = NULL;
166 int ncuts;
167
168 SCIP_CALL( SCIPallocBufferArray(scip, &cuts, graph->nnodes) );
169 for(int i = 0; i < graph->nnodes; i++)
170 {
171 SCIP_CALL( SCIPallocBufferArray(scip, &cuts[i], graph->nnodes) ); /*lint !e866*/
172 }
173
174 // try to find cuts
175 if( ghc_tree(graph, cuts, &ncuts, SCIPfeastol(scip)) )
176 {
177 // create a new cutting plane for every suitable arc (representing a cut with value < 2) of the Gomory Hu Tree
178 for(int i = 0; i < ncuts && *result != SCIP_CUTOFF; ++i)
179 {
180 SCIP_ROW* row;
181 SCIP_CALL( SCIPcreateEmptyRowConshdlr(scip, &row, conshdlr, "subtour", 2.0, SCIPinfinity(scip), FALSE, FALSE, TRUE) );
182
184
185 for(int j = 0; j < graph->nnodes; j++)
186 {
187 // in gmincut the graph has been partitioned into two parts, represented by bools
188 if( cuts[i][j] )
189 {
190 GRAPHEDGE* edge = graph->nodes[j].first_edge;
191
192 // take every edge with nodes in different parts into account
193 while( edge != NULL )
194 {
195 if( ! cuts[i][edge->adjac->id] )
196 {
197 SCIP_CALL( SCIPaddVarToRow(scip, row, edge->var, 1.0) );
198 }
199 edge = edge->next;
200 }
201 }
202 }
203
205
206 // Add violated cut. The cuts produced by ghc_tree are violated by at least the feasibility tolerance. If we
207 // are enforcing, then this is enough to add the cut. Otherwise (we are separating), we check whether the
208 // cut is efficacious.
209 if( enforce || SCIPisCutEfficacious(scip, sol, row) )
210 {
211 SCIP_Bool infeasible;
212 SCIP_CALL( SCIPaddRow(scip, row, FALSE, &infeasible) );
213 if ( infeasible )
214 *result = SCIP_CUTOFF;
215 else
216 *result = SCIP_SEPARATED;
217 }
218 SCIP_CALL( SCIPreleaseRow(scip, &row) );
219 }
220 }
221
222 for(int i = graph->nnodes - 1; i >= 0; i--)
223 SCIPfreeBufferArray(scip, &cuts[i]);
225 }
226
227 return SCIP_OKAY;
228}
229
230
231/** frees specific constraint data */
232SCIP_DECL_CONSDELETE(ConshdlrSubtour::scip_delete)
233{ /*lint --e{715}*/
234 assert(consdata != NULL);
235
236 release_graph(&(*consdata)->graph);
237 SCIPfreeBlockMemory(scip, consdata);
238
239 return SCIP_OKAY;
240}
241
242
243/** transforms constraint data into data belonging to the transformed problem */
244SCIP_DECL_CONSTRANS(ConshdlrSubtour::scip_trans)
245{
246 SCIP_CONSDATA* sourcedata;
247 SCIP_CONSDATA* targetdata = NULL;
248
249 sourcedata = SCIPconsGetData(sourcecons);
250 assert( sourcedata != NULL );
251
252 SCIP_CALL( SCIPallocBlockMemory(scip, &targetdata) );
253 targetdata->graph = sourcedata->graph;
254 capture_graph(targetdata->graph);
255
256 /* create target constraint */
257 SCIP_CALL( SCIPcreateCons(scip, targetcons, SCIPconsGetName(sourcecons), conshdlr, targetdata,
258 SCIPconsIsInitial(sourcecons), SCIPconsIsSeparated(sourcecons), SCIPconsIsEnforced(sourcecons),
259 SCIPconsIsChecked(sourcecons), SCIPconsIsPropagated(sourcecons), SCIPconsIsLocal(sourcecons),
260 SCIPconsIsModifiable(sourcecons), SCIPconsIsDynamic(sourcecons), SCIPconsIsRemovable(sourcecons),
261 SCIPconsIsStickingAtNode(sourcecons)) );
262
263 return SCIP_OKAY;
264}
265
266
267/** separation method of constraint handler for LP solution
268 *
269 * Separates all constraints of the constraint handler. The method is called in the LP solution loop,
270 * which means that a valid LP solution exists.
271 *
272 * The first nusefulconss constraints are the ones, that are identified to likely be violated. The separation
273 * method should process only the useful constraints in most runs, and only occasionally the remaining
274 * nconss - nusefulconss constraints.
275 *
276 * possible return values for *result (if more than one applies, the first in the list should be used):
277 * - SCIP_CUTOFF : the node is infeasible in the variable's bounds and can be cut off
278 * - SCIP_CONSADDED : an additional constraint was generated
279 * - SCIP_REDUCEDDOM : a variable's domain was reduced
280 * - SCIP_SEPARATED : a cutting plane was generated
281 * - SCIP_DIDNOTFIND : the separator searched, but did not find domain reductions, cutting planes, or cut constraints
282 * - SCIP_DIDNOTRUN : the separator was skipped
283 * - SCIP_DELAYED : the separator was skipped, but should be called again
284 */
285SCIP_DECL_CONSSEPALP(ConshdlrSubtour::scip_sepalp)
286{
287 SCIP_CALL( sepaSubtour(scip, conshdlr, conss, nconss, nusefulconss, NULL, FALSE, result) );
288
289 return SCIP_OKAY;
290}
291
292
293/** separation method of constraint handler for arbitrary primal solution
294 *
295 * Separates all constraints of the constraint handler. The method is called outside the LP solution loop (e.g., by
296 * a relaxator or a primal heuristic), which means that there is no valid LP solution.
297 * Instead, the method should produce cuts that separate the given solution.
298 *
299 * The first nusefulconss constraints are the ones, that are identified to likely be violated. The separation
300 * method should process only the useful constraints in most runs, and only occasionally the remaining
301 * nconss - nusefulconss constraints.
302 *
303 * possible return values for *result (if more than one applies, the first in the list should be used):
304 * - SCIP_CUTOFF : the node is infeasible in the variable's bounds and can be cut off
305 * - SCIP_CONSADDED : an additional constraint was generated
306 * - SCIP_REDUCEDDOM : a variable's domain was reduced
307 * - SCIP_SEPARATED : a cutting plane was generated
308 * - SCIP_DIDNOTFIND : the separator searched, but did not find domain reductions, cutting planes, or cut constraints
309 * - SCIP_DIDNOTRUN : the separator was skipped
310 * - SCIP_DELAYED : the separator was skipped, but should be called again
311 */
312SCIP_DECL_CONSSEPASOL(ConshdlrSubtour::scip_sepasol)
313{
314 SCIP_CALL( sepaSubtour(scip, conshdlr, conss, nconss, nusefulconss, sol, FALSE, result) );
315
316 return SCIP_OKAY;
317}
318
319
320/** constraint enforcing method of constraint handler for LP solutions
321 *
322 * The method is called at the end of the node processing loop for a node where the LP was solved.
323 * The LP solution has to be checked for feasibility. If possible, an infeasibility should be resolved by
324 * branching, reducing a variable's domain to exclude the solution or separating the solution with a valid
325 * cutting plane.
326 *
327 * The enforcing methods of the active constraint handlers are called in decreasing order of their enforcing
328 * priorities until the first constraint handler returned with the value SCIP_CUTOFF, SCIP_SEPARATED,
329 * SCIP_REDUCEDDOM, SCIP_CONSADDED, or SCIP_BRANCHED.
330 * The integrality constraint handler has an enforcing priority of zero. A constraint handler which can
331 * (or wants) to enforce its constraints only for integral solutions should have a negative enforcing priority
332 * (e.g. the alldiff-constraint can only operate on integral solutions).
333 * A constraint handler which wants to incorporate its own branching strategy even on non-integral
334 * solutions must have an enforcing priority greater than zero (e.g. the SOS-constraint incorporates
335 * SOS-branching on non-integral solutions).
336 *
337 * The first nusefulconss constraints are the ones, that are identified to likely be violated. The enforcing
338 * method should process the useful constraints first. The other nconss - nusefulconss constraints should only
339 * be enforced, if no violation was found in the useful constraints.
340 *
341 * possible return values for *result (if more than one applies, the first in the list should be used):
342 * - SCIP_CUTOFF : the node is infeasible in the variable's bounds and can be cut off
343 * - SCIP_CONSADDED : an additional constraint was generated
344 * - SCIP_REDUCEDDOM : a variable's domain was reduced
345 * - SCIP_SEPARATED : a cutting plane was generated
346 * - SCIP_BRANCHED : no changes were made to the problem, but a branching was applied to resolve an infeasibility
347 * - SCIP_INFEASIBLE : at least one constraint is infeasible, but it was not resolved
348 * - SCIP_FEASIBLE : all constraints of the handler are feasible
349 */
350SCIP_DECL_CONSENFOLP(ConshdlrSubtour::scip_enfolp)
351{ /*lint --e{715}*/
352 assert(result != NULL);
353
354 // search for subtour elimination cuts
355 SCIP_CALL( sepaSubtour(scip, conshdlr, conss, nconss, nusefulconss, NULL, TRUE, result) );
356
357 // if separation could not find a subtour, then the solution is feasible for this constraint
358 if( *result == SCIP_DIDNOTFIND )
359 *result = SCIP_FEASIBLE;
360
361 return SCIP_OKAY;
362}
363
364/** constraint enforcing method of constraint handler for pseudo solutions
365 *
366 * The method is called at the end of the node processing loop for a node where the LP was not solved.
367 * The pseudo solution has to be checked for feasibility. If possible, an infeasibility should be resolved by
368 * branching, reducing a variable's domain to exclude the solution or adding an additional constraint.
369 * Separation is not possible, since the LP is not processed at the current node. All LP informations like
370 * LP solution, slack values, or reduced costs are invalid and must not be accessed.
371 *
372 * Like in the enforcing method for LP solutions, the enforcing methods of the active constraint handlers are
373 * called in decreasing order of their enforcing priorities until the first constraint handler returned with
374 * the value SCIP_CUTOFF, SCIP_REDUCEDDOM, SCIP_CONSADDED, SCIP_BRANCHED, or SCIP_SOLVELP.
375 *
376 * The first nusefulconss constraints are the ones, that are identified to likely be violated. The enforcing
377 * method should process the useful constraints first. The other nconss - nusefulconss constraints should only
378 * be enforced, if no violation was found in the useful constraints.
379 *
380 * If the pseudo solution's objective value is lower than the lower bound of the node, it cannot be feasible
381 * and the enforcing method may skip it's check and set *result to SCIP_DIDNOTRUN. However, it can also process
382 * its constraints and return any other possible result code.
383 *
384 * possible return values for *result (if more than one applies, the first in the list should be used):
385 * - SCIP_CUTOFF : the node is infeasible in the variable's bounds and can be cut off
386 * - SCIP_CONSADDED : an additional constraint was generated
387 * - SCIP_REDUCEDDOM : a variable's domain was reduced
388 * - SCIP_BRANCHED : no changes were made to the problem, but a branching was applied to resolve an infeasibility
389 * - SCIP_SOLVELP : at least one constraint is infeasible, and this can only be resolved by solving the SCIP_LP
390 * - SCIP_INFEASIBLE : at least one constraint is infeasible, but it was not resolved
391 * - SCIP_FEASIBLE : all constraints of the handler are feasible
392 * - SCIP_DIDNOTRUN : the enforcement was skipped (only possible, if objinfeasible is true)
393 */
394SCIP_DECL_CONSENFOPS(ConshdlrSubtour::scip_enfops)
395{ /*lint --e{715}*/
396 assert(result != NULL);
397
398 // search for subtour elimination cuts
399 SCIP_CALL( sepaSubtour(scip, conshdlr, conss, nconss, nusefulconss, NULL, TRUE, result) );
400
401 // if separation could not find a subtour, then the solution is feasible for this constraint
402 if( *result == SCIP_DIDNOTFIND )
403 *result = SCIP_FEASIBLE;
404
405 return SCIP_OKAY;
406}
407
408/** feasibility check method of constraint handler for primal solutions
409 *
410 * The given solution has to be checked for feasibility.
411 *
412 * The check methods of the active constraint handlers are called in decreasing order of their check
413 * priorities until the first constraint handler returned with the result SCIP_INFEASIBLE.
414 * The integrality constraint handler has a check priority of zero. A constraint handler which can
415 * (or wants) to check its constraints only for integral solutions should have a negative check priority
416 * (e.g. the alldiff-constraint can only operate on integral solutions).
417 * A constraint handler which wants to check feasibility even on non-integral solutions must have a
418 * check priority greater than zero (e.g. if the check is much faster than testing all variables for
419 * integrality).
420 *
421 * In some cases, integrality conditions or rows of the current LP don't have to be checked, because their
422 * feasibility is already checked or implicitly given. In these cases, 'checkintegrality' or
423 * 'checklprows' is FALSE.
424 *
425 * possible return values for *result:
426 * - SCIP_INFEASIBLE : at least one constraint of the handler is infeasible
427 * - SCIP_FEASIBLE : all constraints of the handler are feasible
428 */
429SCIP_DECL_CONSCHECK(ConshdlrSubtour::scip_check)
430{ /*lint --e{715}*/
431 assert(result != NULL);
432 *result = SCIP_FEASIBLE;
433
434 for( int i = 0; i < nconss; ++i )
435 {
436 SCIP_CONSDATA* consdata;
437 GRAPH* graph;
438 SCIP_Bool found;
439
440 assert(conss != NULL);
441 assert(conss[i] != NULL);
442 consdata = SCIPconsGetData(conss[i]);
443 assert(consdata != NULL);
444 graph = consdata->graph;
445 assert(graph != NULL);
446
447 // if a subtour is found, the solution must be infeasible
448 found = findSubtour(scip, graph, sol);
449 if( found )
450 {
451 *result = SCIP_INFEASIBLE;
452 if( printreason )
453 {
454 SCIP_CALL( SCIPprintCons(scip, conss[i], NULL) );
455 SCIPinfoMessage(scip, NULL, "violation: graph has a subtour\n");
456 }
457 }
458 }
459
460 return SCIP_OKAY;
461}
462
463/** domain propagation method of constraint handler
464 *
465 * The first nusefulconss constraints are the ones, that are identified to likely be violated. The propagation
466 * method should process only the useful constraints in most runs, and only occasionally the remaining
467 * nconss - nusefulconss constraints.
468 *
469 * possible return values for *result:
470 * - SCIP_CUTOFF : the node is infeasible in the variable's bounds and can be cut off
471 * - SCIP_REDUCEDDOM : at least one domain reduction was found
472 * - SCIP_DIDNOTFIND : the propagator searched, but did not find any domain reductions
473 * - SCIP_DIDNOTRUN : the propagator was skipped
474 * - SCIP_DELAYED : the propagator was skipped, but should be called again
475 */
476SCIP_DECL_CONSPROP(ConshdlrSubtour::scip_prop)
477{ /*lint --e{715}*/
478 assert(result != NULL);
479
480 *result = SCIP_DIDNOTRUN;
481 return SCIP_OKAY;
482}
483
484/** variable rounding lock method of constraint handler
485 *
486 * This method is called, after a constraint is added or removed from the transformed problem.
487 * It should update the rounding locks of all associated variables with calls to SCIPaddVarLocksType(),
488 * depending on the way, the variable is involved in the constraint:
489 * - If the constraint may get violated by decreasing the value of a variable, it should call
490 * SCIPaddVarLocksType(scip, var, SCIP_LOCKTYPE_MODEL, nlockspos, nlocksneg), saying that rounding down is
491 * potentially rendering the (positive) constraint infeasible and rounding up is potentially rendering the
492 * negation of the constraint infeasible.
493 * - If the constraint may get violated by increasing the value of a variable, it should call
494 * SCIPaddVarLocksType(scip, var, SCIP_LOCKTYPE_MODEL, nlocksneg, nlockspos), saying that rounding up is
495 * potentially rendering the constraint's negation infeasible and rounding up is potentially rendering the
496 * constraint itself infeasible.
497 * - If the constraint may get violated by changing the variable in any direction, it should call
498 * SCIPaddVarLocksType(scip, var, SCIP_LOCKTYPE_MODEL, nlockspos + nlocksneg, nlockspos + nlocksneg).
499 *
500 * Consider the linear constraint "3x -5y +2z <= 7" as an example. The variable rounding lock method of the
501 * linear constraint handler should call SCIPaddVarLocksType(scip, x, SCIP_LOCKTYPE_MODEL, nlocksneg, nlockspos),
502 * SCIPaddVarLocksType(scip, y, SCIP_LOCKTYPE_MODEL, nlockspos, nlocksneg) and
503 * SCIPaddVarLocksType(scip, z, SCIP_LOCKTYPE_MODEL, nlocksneg, nlockspos) to tell SCIP,
504 * that rounding up of x and z and rounding down of y can destroy the feasibility of the constraint, while rounding
505 * down of x and z and rounding up of y can destroy the feasibility of the constraint's negation "3x -5y +2z > 7".
506 * A linear constraint "2 <= 3x -5y +2z <= 7" should call
507 * SCIPaddVarLocksType(scip, ..., SCIP_LOCKTYPE_MODEL, nlockspos + nlocksneg, nlockspos + nlocksneg) on all variables,
508 * since rounding in both directions of each variable can destroy both the feasibility of the constraint and it's negation
509 * "3x -5y +2z < 2 or 3x -5y +2z > 7".
510 *
511 * If the constraint itself contains other constraints as sub constraints (e.g. the "or" constraint concatenation
512 * "c(x) or d(x)"), the rounding lock methods of these constraints should be called in a proper way.
513 * - If the constraint may get violated by the violation of the sub constraint c, it should call
514 * SCIPaddConsLocks(scip, c, nlockspos, nlocksneg), saying that infeasibility of c may lead to infeasibility of
515 * the (positive) constraint, and infeasibility of c's negation (i.e. feasibility of c) may lead to infeasibility
516 * of the constraint's negation (i.e. feasibility of the constraint).
517 * - If the constraint may get violated by the feasibility of the sub constraint c, it should call
518 * SCIPaddConsLocks(scip, c, nlocksneg, nlockspos), saying that infeasibility of c may lead to infeasibility of
519 * the constraint's negation (i.e. feasibility of the constraint), and infeasibility of c's negation (i.e. feasibility
520 * of c) may lead to infeasibility of the (positive) constraint.
521 * - If the constraint may get violated by any change in the feasibility of the sub constraint c, it should call
522 * SCIPaddConsLocks(scip, c, nlockspos + nlocksneg, nlockspos + nlocksneg).
523 *
524 * Consider the or concatenation "c(x) or d(x)". The variable rounding lock method of the or constraint handler
525 * should call SCIPaddConsLocks(scip, c, nlockspos, nlocksneg) and SCIPaddConsLocks(scip, d, nlockspos, nlocksneg)
526 * to tell SCIP, that infeasibility of c and d can lead to infeasibility of "c(x) or d(x)".
527 *
528 * As a second example, consider the equivalence constraint "y <-> c(x)" with variable y and constraint c. The
529 * constraint demands, that y == 1 if and only if c(x) is satisfied. The variable lock method of the corresponding
530 * constraint handler should call SCIPaddVarLocksType(scip, y, SCIP_LOCKTYPE_MODEL, nlockspos + nlocksneg, nlockspos + nlocksneg) and
531 * SCIPaddConsLocks(scip, c, nlockspos + nlocksneg, nlockspos + nlocksneg), because any modification to the
532 * value of y or to the feasibility of c can alter the feasibility of the equivalence constraint.
533 */
534SCIP_DECL_CONSLOCK(ConshdlrSubtour::scip_lock)
535{ /*lint --e{715}*/
536 SCIP_CONSDATA* consdata;
537 GRAPH* g;
538
539 consdata = SCIPconsGetData(cons);
540 assert(consdata != NULL);
541
542 g = consdata->graph;
543 assert(g != NULL);
544
545 for( int i = 0; i < g->nedges; ++i )
546 {
547 SCIP_CALL( SCIPaddVarLocksType(scip, g->edges[i].var, SCIP_LOCKTYPE_MODEL, nlocksneg, nlockspos) );
548 }
549
550 return SCIP_OKAY;
551}
552
553/** variable deletion method of constraint handler
554 *
555 * This method should iterate over all constraints of the constraint handler and delete all variables
556 * that were marked for deletion by SCIPdelVar().
557 *
558 * input:
559 * - scip : SCIP main data structure
560 * - conshdlr : the constraint handler itself
561 * - conss : array of constraints in transformed problem
562 * - nconss : number of constraints in transformed problem
563 */
564SCIP_DECL_CONSDELVARS(ConshdlrSubtour::scip_delvars)
565{ /*lint --e{715}*/
566 return SCIP_OKAY;
567}
568
569
570/** constraint display method of constraint handler
571 *
572 * The constraint handler should store a representation of the constraint into the given text file.
573 */
574SCIP_DECL_CONSPRINT(ConshdlrSubtour::scip_print)
575{ /*lint --e{715}*/
576 SCIP_CONSDATA* consdata;
577 GRAPH* g;
578
579 consdata = SCIPconsGetData(cons);
580 assert(consdata != NULL);
581
582 g = consdata->graph;
583 assert(g != NULL);
584
585 SCIPinfoMessage(scip, file, "subtour of Graph G with %d nodes and %d edges\n", g->nnodes, g->nedges);
586
587 return SCIP_OKAY;
588}
589
590/** clone method which will be used to copy a objective plugin */
591SCIP_DECL_CONSHDLRCLONE(ObjProbCloneable* ConshdlrSubtour::clone) /*lint !e665*/
592{
593 assert(valid != NULL);
594 *valid = TRUE;
595 return new ConshdlrSubtour(scip);
596}
597
598/** constraint copying method of constraint handler
599 *
600 * The constraint handler can provide a copy method, which copies a constraint from one SCIP data structure into a other
601 * SCIP data structure.
602 */
603SCIP_DECL_CONSCOPY(ConshdlrSubtour::scip_copy)
604{ /*lint --e{715}*/
605 SCIP_CONSHDLR* conshdlr;
606 SCIP_CONSDATA* consdata = NULL;
607
608 assert(valid != NULL);
609
610 /* find the subtour constraint handler */
611 conshdlr = SCIPfindConshdlr(scip, "subtour");
612 if( conshdlr == NULL )
613 {
614 SCIPerrorMessage("subtour constraint handler not found\n");
615 return SCIP_PLUGINNOTFOUND;
616 }
617
618 /* create constraint data */
619 SCIP_CALL( SCIPallocBlockMemory( scip, &consdata) );
620 ProbDataTSP* probdatatsp;
621 probdatatsp = dynamic_cast<ProbDataTSP*>(SCIPgetObjProbData(scip));
622 assert( probdatatsp != NULL );
623
624 GRAPH * graph = probdatatsp->getGraph();
625 consdata->graph = graph;
626 capture_graph( consdata->graph );
627
628 /* create constraint */
629 SCIP_CALL( SCIPcreateCons(scip, cons, (name == NULL) ? SCIPconsGetName(sourcecons) : name,
630 conshdlr, consdata, initial, separate, enforce, check,
631 propagate, local, modifiable, dynamic, removable, FALSE) );
632
633 *valid = TRUE;
634
635 return SCIP_OKAY;
636}
637
638/** creates and captures a TSP subtour constraint */
640 SCIP* scip, /**< SCIP data structure */
641 SCIP_CONS** cons, /**< pointer to hold the created constraint */
642 const char* name, /**< name of constraint */
643 GRAPH* graph, /**< the underlying graph */
644 SCIP_Bool initial, /**< should the LP relaxation of constraint be in the initial LP? */
645 SCIP_Bool separate, /**< should the constraint be separated during LP processing? */
646 SCIP_Bool enforce, /**< should the constraint be enforced during node processing? */
647 SCIP_Bool check, /**< should the constraint be checked for feasibility? */
648 SCIP_Bool propagate, /**< should the constraint be propagated during node processing? */
649 SCIP_Bool local, /**< is constraint only valid locally? */
650 SCIP_Bool modifiable, /**< is constraint modifiable (subject to column generation)? */
651 SCIP_Bool dynamic, /**< is constraint dynamic? */
652 SCIP_Bool removable /**< should the constraint be removed from the LP due to aging or cleanup? */
653 )
654{
655 SCIP_CONSHDLR* conshdlr;
656 SCIP_CONSDATA* consdata;
657
658 /* find the subtour constraint handler */
659 conshdlr = SCIPfindConshdlr(scip, "subtour");
660 if( conshdlr == NULL )
661 {
662 SCIPerrorMessage("subtour constraint handler not found\n");
663 return SCIP_PLUGINNOTFOUND;
664 }
665
666 /* create constraint data */
667 SCIP_CALL( SCIPallocBlockMemory(scip, &consdata) ); /*lint !e530*/
668 consdata->graph = graph;
669 capture_graph(consdata->graph);
670
671 /* create constraint */
672 SCIP_CALL( SCIPcreateCons(scip, cons, name, conshdlr, consdata, initial, separate, enforce, check, propagate,
673 local, modifiable, dynamic, removable, FALSE) );
674
675 return SCIP_OKAY;
676}
SCIP_DECL_CONSENFOPS(ConshdlrSubtour::scip_enfops)
SCIP_DECL_CONSSEPASOL(ConshdlrSubtour::scip_sepasol)
SCIP_DECL_CONSLOCK(ConshdlrSubtour::scip_lock)
static SCIP_Bool findSubtour(SCIP *scip, GRAPH *graph, SCIP_SOL *sol)
SCIP_DECL_CONSHDLRCLONE(ObjProbCloneable *ConshdlrSubtour::clone)
SCIP_DECL_CONSDELETE(ConshdlrSubtour::scip_delete)
static SCIP_RETCODE sepaSubtour(SCIP *scip, SCIP_CONSHDLR *conshdlr, SCIP_CONS **conss, int nconss, int nusefulconss, SCIP_SOL *sol, SCIP_Bool enforce, SCIP_RESULT *result)
SCIP_DECL_CONSCOPY(ConshdlrSubtour::scip_copy)
SCIP_DECL_CONSSEPALP(ConshdlrSubtour::scip_sepalp)
SCIP_DECL_CONSTRANS(ConshdlrSubtour::scip_trans)
SCIP_DECL_CONSDELVARS(ConshdlrSubtour::scip_delvars)
SCIP_DECL_CONSPRINT(ConshdlrSubtour::scip_print)
SCIP_DECL_CONSCHECK(ConshdlrSubtour::scip_check)
SCIP_DECL_CONSPROP(ConshdlrSubtour::scip_prop)
SCIP_DECL_CONSENFOLP(ConshdlrSubtour::scip_enfolp)
C++ constraint handler for TSP subtour elimination constraints.
void capture_graph(GRAPH *gr)
SCIP_Bool ghc_tree(GRAPH *gr, SCIP_Bool **cuts, int *ncuts, double minviol)
void release_graph(GRAPH **gr)
generator for global cuts in undirected graphs
GRAPH * getGraph()
Definition: ProbDataTSP.h:97
Constraint handler for linear constraints in their most general form, .
#define NULL
Definition: def.h:266
#define SCIP_Bool
Definition: def.h:91
#define TRUE
Definition: def.h:93
#define FALSE
Definition: def.h:94
#define SCIP_CALL(x)
Definition: def.h:373
#define nnodes
Definition: gastrans.c:74
void SCIPinfoMessage(SCIP *scip, FILE *file, const char *formatstr,...)
Definition: scip_message.c:208
SCIP_CONSHDLR * SCIPfindConshdlr(SCIP *scip, const char *name)
Definition: scip_cons.c:941
SCIP_CONSDATA * SCIPconsGetData(SCIP_CONS *cons)
Definition: cons.c:8244
SCIP_Bool SCIPconsIsDynamic(SCIP_CONS *cons)
Definition: cons.c:8473
SCIP_Bool SCIPconsIsInitial(SCIP_CONS *cons)
Definition: cons.c:8383
SCIP_RETCODE SCIPprintCons(SCIP *scip, SCIP_CONS *cons, FILE *file)
Definition: scip_cons.c:2537
SCIP_Bool SCIPconsIsChecked(SCIP_CONS *cons)
Definition: cons.c:8413
SCIP_Bool SCIPconsIsEnforced(SCIP_CONS *cons)
Definition: cons.c:8403
SCIP_RETCODE SCIPcreateCons(SCIP *scip, SCIP_CONS **cons, const char *name, SCIP_CONSHDLR *conshdlr, SCIP_CONSDATA *consdata, SCIP_Bool initial, SCIP_Bool separate, SCIP_Bool enforce, SCIP_Bool check, SCIP_Bool propagate, SCIP_Bool local, SCIP_Bool modifiable, SCIP_Bool dynamic, SCIP_Bool removable, SCIP_Bool stickingatnode)
Definition: scip_cons.c:998
SCIP_Bool SCIPconsIsPropagated(SCIP_CONS *cons)
Definition: cons.c:8433
SCIP_Bool SCIPconsIsLocal(SCIP_CONS *cons)
Definition: cons.c:8453
const char * SCIPconsGetName(SCIP_CONS *cons)
Definition: cons.c:8214
SCIP_Bool SCIPconsIsModifiable(SCIP_CONS *cons)
Definition: cons.c:8463
SCIP_Bool SCIPconsIsStickingAtNode(SCIP_CONS *cons)
Definition: cons.c:8493
SCIP_Bool SCIPconsIsSeparated(SCIP_CONS *cons)
Definition: cons.c:8393
SCIP_Bool SCIPconsIsRemovable(SCIP_CONS *cons)
Definition: cons.c:8483
SCIP_Bool SCIPisCutEfficacious(SCIP *scip, SCIP_SOL *sol, SCIP_ROW *cut)
Definition: scip_cut.c:117
SCIP_RETCODE SCIPaddRow(SCIP *scip, SCIP_ROW *row, SCIP_Bool forcecut, SCIP_Bool *infeasible)
Definition: scip_cut.c:250
#define SCIPallocBufferArray(scip, ptr, num)
Definition: scip_mem.h:124
#define SCIPfreeBufferArray(scip, ptr)
Definition: scip_mem.h:136
#define SCIPfreeBlockMemory(scip, ptr)
Definition: scip_mem.h:108
#define SCIPallocBlockMemory(scip, ptr)
Definition: scip_mem.h:89
SCIP_RETCODE SCIPcacheRowExtensions(SCIP *scip, SCIP_ROW *row)
Definition: scip_lp.c:1635
SCIP_RETCODE SCIPflushRowExtensions(SCIP *scip, SCIP_ROW *row)
Definition: scip_lp.c:1658
SCIP_RETCODE SCIPcreateEmptyRowConshdlr(SCIP *scip, SCIP_ROW **row, SCIP_CONSHDLR *conshdlr, const char *name, SCIP_Real lhs, SCIP_Real rhs, SCIP_Bool local, SCIP_Bool modifiable, SCIP_Bool removable)
Definition: scip_lp.c:1391
SCIP_RETCODE SCIPaddVarToRow(SCIP *scip, SCIP_ROW *row, SCIP_VAR *var, SCIP_Real val)
Definition: scip_lp.c:1701
SCIP_RETCODE SCIPreleaseRow(SCIP *scip, SCIP_ROW **row)
Definition: scip_lp.c:1562
SCIP_Real SCIPgetSolVal(SCIP *scip, SCIP_SOL *sol, SCIP_VAR *var)
Definition: scip_sol.c:1213
SCIP_Real SCIPinfinity(SCIP *scip)
SCIP_Real SCIPfeastol(SCIP *scip)
SCIP_RETCODE SCIPaddVarLocksType(SCIP *scip, SCIP_VAR *var, SCIP_LOCKTYPE locktype, int nlocksdown, int nlocksup)
Definition: scip_var.c:4382
Definition: pqueue.h:38
SCIP_RETCODE SCIPcreateConsSubtour(SCIP *scip, SCIP_CONS **cons, const char *name, GRAPH *graph, SCIP_Bool initial, SCIP_Bool separate, SCIP_Bool enforce, SCIP_Bool check, SCIP_Bool propagate, SCIP_Bool local, SCIP_Bool modifiable, SCIP_Bool dynamic, SCIP_Bool removable)
scip::ObjProbData * SCIPgetObjProbData(SCIP *scip)
C++ wrapper classes for SCIP.
#define SCIPerrorMessage
Definition: pub_message.h:64
struct GraphEdge * back
Definition: GomoryHuTree.h:70
double rcap
Definition: GomoryHuTree.h:66
double cap
Definition: GomoryHuTree.h:65
GRAPHNODE * adjac
Definition: GomoryHuTree.h:72
SCIP_VAR * var
Definition: GomoryHuTree.h:74
struct GraphEdge * next
Definition: GomoryHuTree.h:69
struct GraphEdge * first_edge
Definition: GomoryHuTree.h:53
GRAPHNODE * nodes
Definition: GomoryHuTree.h:86
int nedges
Definition: GomoryHuTree.h:83
int nnodes
Definition: GomoryHuTree.h:82
GRAPHEDGE * edges
Definition: GomoryHuTree.h:87
Definition of base class for all clonable classes which define problem data.
struct SCIP_ConsData SCIP_CONSDATA
Definition: type_cons.h:65
@ SCIP_DIDNOTRUN
Definition: type_result.h:42
@ SCIP_CUTOFF
Definition: type_result.h:48
@ SCIP_FEASIBLE
Definition: type_result.h:45
@ SCIP_DIDNOTFIND
Definition: type_result.h:44
@ SCIP_SEPARATED
Definition: type_result.h:49
@ SCIP_INFEASIBLE
Definition: type_result.h:46
enum SCIP_Result SCIP_RESULT
Definition: type_result.h:61
@ SCIP_PLUGINNOTFOUND
Definition: type_retcode.h:54
@ SCIP_OKAY
Definition: type_retcode.h:42
enum SCIP_Retcode SCIP_RETCODE
Definition: type_retcode.h:63
@ SCIP_LOCKTYPE_MODEL
Definition: type_var.h:97