org.chocosolver.solver.constraints.nary.knapsack.PropKnapsackKatriel01 Maven / Gradle / Ivy
/*
* This file is part of choco-solver, http://choco-solver.org/
*
* Copyright (c) 2022, IMT Atlantique. All rights reserved.
*
* Licensed under the BSD 4-clause license.
*
* See LICENSE file in the project root for full license information.
*/
package org.chocosolver.solver.constraints.nary.knapsack;
import org.chocosolver.memory.IEnvironment;
import org.chocosolver.solver.constraints.Propagator;
import org.chocosolver.solver.constraints.PropagatorPriority;
import org.chocosolver.solver.constraints.nary.knapsack.structure.*;
import org.chocosolver.solver.exception.ContradictionException;
import org.chocosolver.solver.variables.BoolVar;
import org.chocosolver.solver.variables.IntVar;
import org.chocosolver.solver.variables.events.IntEventType;
import org.chocosolver.solver.variables.events.PropagatorEventType;
import org.chocosolver.util.ESat;
import org.chocosolver.util.sort.ArraySort;
import org.chocosolver.util.tools.ArrayUtils;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.LinkedList;
import java.util.List;
/**
* Propagator for the 0/1-Knapsack constraint
* based on Dantzig-Wolfe relaxation trying
* to find forbidden and mandatory items
*
* @author Nicolas PIERRE
*/
public class PropKnapsackKatriel01 extends Propagator {
static final int ADDED = 1;
static final int REMOVED = -1;
static final int NOT_DEFINED = 0;
// ***********************************************************************************
// VARIABLES
// ***********************************************************************************
private final int[] order;
private final int[] reverseOrder;
private final int n;
private final IntVar capacity;
private final IntVar power;
private final int[] itemState;
private final ItemFindingSearchTree findingTree;
private final ComputingLossWeightTree computingTree;
private Info criticalItemInfos;
// variables to keep track of the change from the original constraint expression
private int usedCapacity;
private int powerCreated;
private boolean mustRecomputeCriticalInfos;
private int lastWorld = -1;
private long lastNbOfBacktracks = -1;
private long lastNbOfRestarts = -1;
// ***********************************************************************************
// CONSTRUCTORS
// ***********************************************************************************
public PropKnapsackKatriel01(BoolVar[] itemOccurence, IntVar capacity, IntVar power,
int[] weight, int[] energy) {
super(ArrayUtils.append(itemOccurence, new IntVar[]{capacity, power}), PropagatorPriority.QUADRATIC, true);
this.n = itemOccurence.length;
this.itemState = new int[n];
this.reverseOrder = new int[n];
this.capacity = capacity;
this.power = power;
this.usedCapacity = 0;
this.powerCreated = 0;
Arrays.fill(this.itemState, 0);
// we find the decreasing order of efficiency
this.order = ArrayUtils.array(0, n - 1);
ArraySort sorter = new ArraySort<>(n, false, true);
sorter.sort(order, n, (i1, i2) -> {
// Compares efficiencies decreasingly
long comparaison = (long) energy[i2] * weight[i1] - (long) energy[i1] * weight[i2];
if (comparaison == 0) {
if (weight[i1] * weight[i2] == 0) {
comparaison = (long) energy[i2] - energy[i1];
} else {
// breaking ties in favor of larger weights
comparaison = (long) weight[i2] - weight[i1];
}
}
return Long.signum(comparaison);
});
ArrayList orderedItems = new ArrayList<>();
orderedItems.ensureCapacity(n);
for (int i = 0; i < n; ++i) {
orderedItems.add(new KPItem(energy[order[i]], weight[order[i]]));
reverseOrder[order[i]] = i;
}
this.findingTree = new ItemFindingSearchTree(orderedItems);
this.computingTree = new ComputingLossWeightTree(orderedItems);
this.mustRecomputeCriticalInfos = true;
}
// ***********************************************************************************
// METHODS
// ***********************************************************************************
@Override
public int getPropagationConditions(int vIdx) {
if (vIdx < n) {
// updates on items
return IntEventType.boundAndInst();
} else if (vIdx == n) {
// updates on the max weight
return IntEventType.combine(IntEventType.DECUPP, IntEventType.INSTANTIATE);
} else /* vIdx == n + 1 */ {
// updates on the energy variable
return IntEventType.combine(IntEventType.INCLOW, IntEventType.INSTANTIATE);
}
}
@Override
public void propagate(int evtmask) throws ContradictionException {
checkWorld();
if (mustRecomputeCriticalInfos()) {
// compute the Dantzig solution and set members variables
this.criticalItemInfos = this.computingTree.findCriticalItem(this.capacity.getUB() - usedCapacity);
mustRecomputeCriticalInfos = false;
}
// if power LB is not reachable so we can't filter
// (every item would be mandatory and forbidden)
if (criticalItemInfos.profit + powerCreated >= power.getLB()) {
List mandatoryList = findMandatoryItems();
List forbiddenList = findForbiddenItems();
for (int unorderedLeafIdx : mandatoryList) {
addItemToSolution(unorderedLeafIdx, true);
// case 3. from mustRecomputeCriticalInfos()
mustRecomputeCriticalInfos = true;
}
for (int unorderedLeafIdx : forbiddenList) {
removeItemFromProblem(unorderedLeafIdx, true);
// case 3. from mustRecomputeCriticalInfos()
mustRecomputeCriticalInfos = true;
}
}
}
@Override
public void propagate(int varIdx, int mask) throws ContradictionException {
if (varIdx < n) {
// item changed
// we update the trees
if (this.vars[varIdx].isInstantiatedTo(0)) {
this.removeItemFromProblem(varIdx, false);
} else if (this.vars[varIdx].isInstantiatedTo(1)) {
this.addItemToSolution(varIdx, false); // TODO
}
}
// case 2. from mustRecomputeCriticalInfos()
mustRecomputeCriticalInfos |= varIdx < n + 1;
forcePropagate(PropagatorEventType.FULL_PROPAGATION);
}
/**
* we recompute the Dantzig solution if :
* - a backtrack occured
* - capacity UB changed
* - an item has been determined
*/
private boolean mustRecomputeCriticalInfos() {
checkWorld();
return mustRecomputeCriticalInfos;
}
private void checkWorld() {
int currentworld = model.getEnvironment().getWorldIndex();
long currentbt = model.getSolver().getBackTrackCount();
long currentrestart = model.getSolver().getRestartCount();
if (currentworld < lastWorld || currentbt != lastNbOfBacktracks || currentrestart > lastNbOfRestarts) {
// case 1. from mustRecomputeCriticalInfos()
mustRecomputeCriticalInfos = true;
}
lastWorld = currentworld;
lastNbOfBacktracks = currentbt;
lastNbOfRestarts = currentrestart;
}
@Override
public ESat isEntailed() {
// sum propagators of the KP contraint define the entailment
return ESat.TRUE;
}
/**
* changes the constraint such that the item is in the solution
* Informs backtrack environment what to do
*
* @param i index of an item in the list given in the constructor
* @param removeVarValue true to remove the value in the variable domain
*/
private void addItemToSolution(int i, boolean removeVarValue) throws ContradictionException {
if (this.itemState[i] == NOT_DEFINED) {
this.itemState[i] = ADDED;
int sortedGlobalIndex = computingTree.leafToGlobalIndex(this.reverseOrder[i]);
computingTree.removeLeaf(sortedGlobalIndex);
findingTree.removeLeaf(sortedGlobalIndex);
if (removeVarValue) {
vars[i].removeValue(0, this);
}
getEnvironment().save(() -> activateItemToProblem(i, ADDED));
// we update intern values
this.usedCapacity += computingTree.getLeaf(sortedGlobalIndex).getActivatedWeight();
this.powerCreated += computingTree.getLeaf(sortedGlobalIndex).getActivatedProfit();
}
}
/**
* changes the constraint such that the item is NOT in the solution
* Informs backtrack environment what to do
*
* @param i index of an item in the list given in the constructor
* @param removeVarValue true to remove the value in the variable domain
*/
private void removeItemFromProblem(int i, boolean removeVarValue) throws ContradictionException {
if (this.itemState[i] == NOT_DEFINED) {
this.itemState[i] = REMOVED;
int sortedGlobalIndex = computingTree.leafToGlobalIndex(this.reverseOrder[i]);
computingTree.removeLeaf(sortedGlobalIndex);
findingTree.removeLeaf(sortedGlobalIndex);
if (removeVarValue) {
vars[i].removeValue(1, this);
}
getEnvironment().save(() -> activateItemToProblem(i, REMOVED));
}
}
/**
* changes the propagator state such that the item is not determined anymore
*
* @param i index of an item in the list given in the constructor
* @param expectedState state of the item when reverting
*/
private void activateItemToProblem(int i, int expectedState) {
if (this.itemState[i] == expectedState) {
int sortedGlobalIndex = computingTree.leafToGlobalIndex(this.reverseOrder[i]);
computingTree.activateLeaf(sortedGlobalIndex);
findingTree.activateLeaf(sortedGlobalIndex);
if (this.itemState[i] == ADDED) {
// we update intern values as the item was added to every solutions
this.usedCapacity -= computingTree.getLeaf(sortedGlobalIndex).getActivatedWeight();
this.powerCreated -= computingTree.getLeaf(sortedGlobalIndex).getActivatedProfit();
}
this.itemState[i] = NOT_DEFINED;
} else if (this.itemState[i] != NOT_DEFINED) {
throw new RuntimeException("the item reverted does not have the expected state");
}
}
/**
* exploits {@code computeLimitWeightMandatory} to find all mandatory items in a
* linear scan
*
* @return indexes (given by the constructor) of mandatory items
*/
private List findMandatoryItems() {
List mandatoryList = new LinkedList<>();
double allowedProfitLoss = criticalItemInfos.profit + powerCreated - power.getLB();
// finding first active item
int index = computingTree.leafToGlobalIndex(0);
if (index != -1) {
if (!computingTree.getLeaf(index).isActive()) {
index = findingTree.findNextRightItem(index, criticalItemInfos.index, 0);
}
// not a trivial KP
int maxWeight = 0;
double criticalItemWeightNotInDantzig = 0;
if (computingTree.isLeaf(criticalItemInfos.index)) {
criticalItemWeightNotInDantzig = computingTree.getNodeWeight(criticalItemInfos.index)
- (criticalItemInfos.weight - criticalItemInfos.weightWithoutCriticalItem);
}
SearchInfos infos = new SearchInfos(false, criticalItemInfos.index,
0, 0, criticalItemWeightNotInDantzig);
while (index != -1) {
infos = computingTree.computeLimitWeightMandatory(criticalItemInfos, index, infos.endItem,
infos.profitAccumulated, infos.weightAccumulated, allowedProfitLoss,
infos.remainingWeightEndItem);
if (infos.decision) {
mandatoryList.add(order[computingTree.globalToLeaf(index)]);
} else {
maxWeight = Math.max(maxWeight, computingTree.getNodeWeight(index));
maxWeight = Math.max(maxWeight, (int) infos.weightAccumulated);
}
index = findingTree.findNextRightItem(index, criticalItemInfos.index, maxWeight);
}
}
return mandatoryList;
}
/**
* exploits {@code computeLimitWeightForbidden} to find all forbidden items in a
* linear scan
*
* @return indexes (given by the constructor) of forbidden items
*/
private List findForbiddenItems() {
List forbiddenList = new LinkedList<>();
double allowedProfitLoss = criticalItemInfos.profit + powerCreated - power.getLB();
// finding first active item
int index = criticalItemInfos.index;
if (index != -1 && criticalItemInfos.index != computingTree.getNumberNodes()) {
int maxWeight = 0;
double criticalItemWeightInDantzig = criticalItemInfos.weight - criticalItemInfos.weightWithoutCriticalItem;
if (!computingTree.getLeaf(index).isActive()) {
index = findingTree.findNextRightItem(index, computingTree.getNumberNodes() - 1, maxWeight);
}
SearchInfos infos = new SearchInfos(false, criticalItemInfos.index,
0, 0, criticalItemWeightInDantzig);
while (index != -1) {
infos = computingTree.computeLimitWeightForbidden(criticalItemInfos, index, infos.endItem,
infos.profitAccumulated, infos.weightAccumulated, allowedProfitLoss,
infos.remainingWeightEndItem);
if (infos.decision) {
forbiddenList.add(order[computingTree.globalToLeaf(index)]);
} else {
maxWeight = Math.max(maxWeight, (int) infos.weightAccumulated);
maxWeight = Math.max(maxWeight, computingTree.getNodeWeight(index));
}
index = findingTree.findNextRightItem(index, computingTree.getNumberNodes() - 1, maxWeight);
}
}
return forbiddenList;
}
private IEnvironment getEnvironment() {
return this.getModel().getEnvironment();
}
}
© 2015 - 2024 Weber Informatics LLC | Privacy Policy