/// <summary>
/// Checks whether the boolean selection is valid w.r.t. the variability model. Does not check for numeric options' correctness.
/// </summary>
/// <param name="config">The list of binary options that are SELECTED (only selected options must occur in the list).</param>
/// <param name="vm">The variability model that represents the context of the configuration.</param>
/// <param name="partialConfiguration">Whether the given list of options represents only a partial configuration. This means that options not in config might be additionally select to obtain a valid configuration.</param>
/// <returns>True if it is a valid selection w.r.t. the VM, false otherwise</returns>
public bool checkConfigurationSAT(List <BinaryOption> config, VariabilityModel vm, bool partialConfiguration)
{
List <CspTerm> variables = new List <CspTerm>();
Dictionary <BinaryOption, CspTerm> elemToTerm = new Dictionary <BinaryOption, CspTerm>();
Dictionary <CspTerm, BinaryOption> termToElem = new Dictionary <CspTerm, BinaryOption>();
ConstraintSystem S = CSPsolver.getConstraintSystem(out variables, out elemToTerm, out termToElem, vm);
//Feature Selection
foreach (BinaryOption binayOpt in elemToTerm.Keys)
{
CspTerm term = elemToTerm[binayOpt];
if (config.Contains(binayOpt))
{
S.AddConstraints(S.Implies(S.True, term));
}
else if (!partialConfiguration)
{
S.AddConstraints(S.Implies(S.True, S.Not(term)));
}
}
ConstraintSolverSolution sol = S.Solve();
if (sol.HasFoundSolution)
{
return(true);
}
else
{
return(false);
}
}
public void Solve(int?[,] field)
{
ConstraintSystem S = ConstraintSystem.CreateSolver();
CspDomain Z = S.CreateIntegerInterval(1, 9);
CspTerm[][] sudoku = S.CreateVariableArray(Z, "cell", 9, 9);
for (int row = 0; row < 9; row++)
{
for (int col = 0; col < 9; col++)
{
if (field[row, col] > 0)
{
S.AddConstraints(S.Equal(field[row, col] ?? 0, sudoku[row][col]));
}
}
S.AddConstraints(S.Unequal(GetSlice(sudoku, row, row, 0, 8)));
}
for (int col = 0; col < 9; col++)
{
S.AddConstraints(S.Unequal(GetSlice(sudoku, 0, 8, col, col)));
}
for (int a = 0; a < 3; a++)
{
for (int b = 0; b < 3; b++)
{
S.AddConstraints(S.Unequal(GetSlice(sudoku, a * 3, a * 3 + 2, b * 3, b * 3 + 2)));
}
}
ConstraintSolverSolution soln = S.Solve();
if (!soln.HasFoundSolution)
{
throw new NoSolutionException("Для поставленной цифры нет решение!");
}
object[] h = new object[9];
for (int row = 0; row < 9; row++)
{
if ((row % 3) == 0)
{
System.Console.WriteLine();
}
for (int col = 0; col < 9; col++)
{
soln.TryGetValue(sudoku[row][col], out h[col]);
}
System.Console.WriteLine("{0}{1}{2} {3}{4}{5} {6}{7}{8}", h[0], h[1], h[2], h[3], h[4], h[5], h[6], h[7], h[8]);
}
}
/// <summary>
/// Checks whether the boolean selection is valid w.r.t. the variability model. Does not check for numeric options' correctness.
/// </summary>
/// <param name="config">The list of binary options that are SELECTED (only selected options must occur in the list).</param>
/// <param name="vm">The variability model that represents the context of the configuration.</param>
/// <param name="exact">Checks also the number of selected options such that it returns only true if exactly the given configuration is valid
/// (e.g., if we need to select more features to get a valid config, it returns false if exact is set to true).
/// <returns>True if it is a valid selection w.r.t. the VM, false otherwise</returns>
public bool checkConfigurationSAT(List <BinaryOption> config, VariabilityModel vm, bool exact)
{
List <CspTerm> variables = new List <CspTerm>();
Dictionary <BinaryOption, CspTerm> elemToTerm = new Dictionary <BinaryOption, CspTerm>();
Dictionary <CspTerm, BinaryOption> termToElem = new Dictionary <CspTerm, BinaryOption>();
ConstraintSystem S = CSPsolver.getConstraintSystem(out variables, out elemToTerm, out termToElem, vm);
//Feature Selection
foreach (BinaryOption binayOpt in elemToTerm.Keys)
{
CspTerm term = elemToTerm[binayOpt];
if (config.Contains(binayOpt))
{
S.AddConstraints(S.Implies(S.True, term));
}
else
{
if (exact)
{
S.AddConstraints(S.Implies(S.True, S.Not(term)));
}
}
}
ConstraintSolverSolution sol = S.Solve();
if (sol.HasFoundSolution)
{
int count = 0;
foreach (CspTerm cT in variables)
{
if (sol.GetIntegerValue(cT) == 1)
{
count++;
}
}
//Needs testing TODO
if (count != config.Count && exact == true)
{
return(false);
}
return(true);
}
else
{
return(false);
}
}
private void AddBinaryConfigurationsToConstraintSystem(VariabilityModel vm, ConstraintSystem s, Configuration configurationToExclude, Dictionary <BinaryOption, CspTerm> elemToTerm)
{
List <BinaryOption> allBinaryOptions = vm.BinaryOptions;
List <CspTerm> positiveTerms = new List <CspTerm>();
List <CspTerm> negativeTerms = new List <CspTerm>();
foreach (BinaryOption binOpt in allBinaryOptions)
{
if (configurationToExclude.BinaryOptions.ContainsKey(binOpt) && configurationToExclude.BinaryOptions[binOpt] == BinaryOption.BinaryValue.Selected)
{
positiveTerms.Add(elemToTerm[binOpt]);
}
else
{
negativeTerms.Add(elemToTerm[binOpt]);
}
}
if (negativeTerms.Count > 0)
{
positiveTerms.Add(s.Not(s.And(negativeTerms.ToArray())));
}
s.AddConstraints(s.Not(s.And(positiveTerms.ToArray())));
}
/// <summary>
/// The method aims at finding a configuration which is similar to the given configuration, but does not contain the optionToBeRemoved. If further options need to be removed from the given configuration, they are outputed in removedElements.
/// Idea: Encode this as a CSP problem. We aim at finding a configuration that maximizes a goal. Each option of the given configuration gets a large value assigned. All other options of the variability model gets a negative value assigned.
/// We will further create a boolean constraint that forbids selecting the optionToBeRemoved. Now, we find an optimal valid configuration.
/// </summary>
/// <param name="optionToBeRemoved">The binary configuration option that must not be part of the new configuration.</param>
/// <param name="originalConfig">The configuration for which we want to find a similar one.</param>
/// <param name="removedElements">If further options need to be removed from the given configuration to build a valid configuration, they are outputed in this list.</param>
/// <param name="vm">The variability model containing all options and their constraints.</param>
/// <returns>A configuration that is valid, similar to the original configuration and does not contain the optionToBeRemoved.</returns>
public List <BinaryOption> GenerateConfigWithoutOption(BinaryOption optionToBeRemoved, List <BinaryOption> originalConfig, out List <BinaryOption> removedElements, VariabilityModel vm)
{
List <CspTerm> variables = new List <CspTerm>();
Dictionary <BinaryOption, CspTerm> elemToTerm = new Dictionary <BinaryOption, CspTerm>();
Dictionary <CspTerm, BinaryOption> termToElem = new Dictionary <CspTerm, BinaryOption>();
ConstraintSystem S = CSPsolver.getConstraintSystem(out variables, out elemToTerm, out termToElem, vm);
removedElements = new List <BinaryOption>();
//Forbid the selection of this configuration option
CspTerm optionToRemove = elemToTerm[optionToBeRemoved];
S.AddConstraints(S.Implies(S.True, S.Not(optionToRemove)));
//Defining Goals
CspTerm[] finalGoals = new CspTerm[variables.Count];
int r = 0;
foreach (var term in variables)
{
if (originalConfig.Contains(termToElem[term]))
{
finalGoals[r] = term * -1000; //Since we minimize, we put a large negative value of an option that is within the original configuration to increase chances that the option gets selected again
}
else
{
finalGoals[r] = variables[r] * 10000;//Positive number will lead to a small chance that an option gets selected when it is not in the original configuration
}
r++;
}
S.TryAddMinimizationGoals(S.Sum(finalGoals));
ConstraintSolverSolution soln = S.Solve();
List <BinaryOption> tempConfig = new List <BinaryOption>();
if (soln.HasFoundSolution && soln.Quality == ConstraintSolverSolution.SolutionQuality.Optimal)
{
tempConfig.Clear();
foreach (CspTerm cT in variables)
{
if (soln.GetIntegerValue(cT) == 1)
{
tempConfig.Add(termToElem[cT]);
}
}
//Adding the options that have been removed from the original configuration
foreach (var opt in originalConfig)
{
if (!tempConfig.Contains(opt))
{
removedElements.Add(opt);
}
}
return(tempConfig);
}
return(null);
}
CspTerm[][] CreateConstrainSystemMatrix(ConstraintSystem system)
{
var size = HouseNumbers.Count;
var matrix = system.CreateBooleanArray(new object(), size, size);
Enumerable.Range(0, size).ToList().ForEach(i =>
{
var row = system.CreateBooleanVector(new object(), size);
var column = system.CreateBooleanVector(new object(), size);
Enumerable.Range(0, size).ToList().ForEach(j =>
{
row[j] = matrix[i][j];
column[j] = matrix[j][i];
});
system.AddConstraints(system.Equal(1, system.Sum(row)));
system.AddConstraints(system.Equal(1, system.Sum(column)));
});
return(matrix);
}
public override void AddConstaint()
{
ConstraintSystem solver = ConstraintSystemSolver.Instance.Solver;
CspTerm constraint = null;
CspTerm inputTerm = Input1.CspTerm;
CspTerm outputTerm = Output.CspTerm;
Type consType = type;
if (IsNotHealthy)
{
// In case the gate is Broken (Not Healthy) - we don't want to add any constraint!!!
return;
/*
* switch (type)
* {
* case Type.buffer:
* consType = Type.not;
* break;
* case Type.not:
* consType = Type.buffer;
* break;
* }
*/
}
lock (ConstraintSystemSolver.Instance.Locker)
{
//Debug.WriteLine("SAT IN!");
switch (consType)
{
case Type.buffer:
constraint = solver.Equal(inputTerm, outputTerm);
break;
case Type.not:
constraint = solver.Equal(inputTerm, solver.Not(outputTerm));
break;
}
solver.AddConstraints(constraint);
//Debug.WriteLine("SAT OUT!");
}
}
static void Main(string[] args)
{
ConstraintSystem s1 = ConstraintSystem.CreateSolver();
// () and ()
CspTerm p1 = s1.CreateBoolean("p1");
CspTerm p2 = s1.CreateBoolean("p2");
CspTerm p3 = s1.CreateBoolean("p3");
CspTerm p4 = s1.CreateBoolean("p4");
var x = new CspDomain();
CspTerm test = s1.And(s1.Or(p1, s1.And(s1.Neg(p3)), s1.Neg(p1)), s1.And(p2, s1.Neg(s1.Difference(p1, p2))));
CspTerm tOr12 = s1.Or(s1.Neg(t1), s1.Neg(t2));
CspTerm tOr13 = s1.Or(s1.Neg(t1), s1.Neg(t3));
CspTerm tOr14 = s1.Or(s1.Neg(t1), s1.Neg(t4));
CspTerm tOr23 = s1.Or(s1.Neg(t2), s1.Neg(t3));
CspTerm tOr24 = s1.Or(s1.Neg(t2), s1.Neg(t4));
CspTerm tOr34 = s1.Or(s1.Neg(t3), s1.Neg(t4));
CspTerm tOr = s1.Or(t1, t2, t3, t4);
s1.AddConstraints(tOr12);
s1.AddConstraints(tOr13);
s1.AddConstraints(tOr14);
s1.AddConstraints(tOr23);
s1.AddConstraints(tOr24);
s1.AddConstraints(tOr34);
s1.AddConstraints(tOr);
ConstraintSolverSolution solution1 = s1.Solve();
if (solution1.HasFoundSolution)
{
Console.WriteLine("Is Satisfiable");
}
else
{
Console.WriteLine("Not satisfiable");
}
Console.ReadKey();
}
private List <List <BinaryOption> > generateTilSize(int i1, int size, int timeout, VariabilityModel vm)
{
var foundSolutions = new List <List <BinaryOption> >();
List <CspTerm> variables = new List <CspTerm>();
Dictionary <BinaryOption, CspTerm> elemToTerm = new Dictionary <BinaryOption, CspTerm>();
Dictionary <CspTerm, BinaryOption> termToElem = new Dictionary <CspTerm, BinaryOption>();
ConstraintSystem S = CSPsolver.getConstraintSystem(out variables, out elemToTerm, out termToElem, vm);
CspTerm t = S.ExactlyMofN(i1, variables.ToArray());
S.AddConstraints(new CspTerm[] { t });
var csp = new ConstraintSolverParams
{
TimeLimitMilliSec = timeout * 1000,
};
ConstraintSolverSolution soln = S.Solve(csp);
int counter = 0;
while (soln.HasFoundSolution)
{
List <BinaryOption> tempConfig = (
from cT
in variables
where soln.GetIntegerValue(cT) == 1
select termToElem[cT]).ToList();
if (tempConfig.Contains(null))
{
tempConfig.Remove(null);
}
foundSolutions.Add(tempConfig);
counter++;
if (counter == size)
{
break;
}
soln.GetNext();
}
//Console.WriteLine(i1 + "\t" + foundSolutions.Count);
return(foundSolutions);
}
/// <summary>
/// Knapsack enumerator -- enumerate up to "numAnswers" combinations of "weights" such that the sum of the weights is less than the weight limit.
/// It places the patterns of items inside the list of patterns. The efficiency parameter ensures that we don't output any which use less than "efficiency" percent
/// off the weightlimit.
/// </summary>
/// <param name="maxAnswers">maximum number of combinations to get out. Limits runtime. If zero return all.</param>
/// <param name="weights">weight of each item to go into the knapsack</param>
/// <param name="weightLimit">knapsack weight limit</param>
/// <param name="efficiency">limit patterns to use at least this % of the weight limit (between 0.0 and 1.0) </param>
/// <param name="patterns">output list of patterns of inclusion of the weights.</param>
public static void SolveKnapsack(int maxAnswers, int[] weights, int weightLimit, double efficiency, out List <int[]> patterns)
{
// convenience value.
int NumItems = weights.Length;
ConstraintSystem solver = ConstraintSystem.CreateSolver();
CspDomain dom = solver.CreateIntegerInterval(0, weightLimit);
CspTerm knapsackSize = solver.Constant(weightLimit);
// these represent the quantity of each item.
CspTerm[] itemQty = solver.CreateVariableVector(dom, "Quantity", NumItems);
CspTerm[] itemWeights = new CspTerm[NumItems];
for (int cnt = 0; cnt < NumItems; cnt++)
{
itemWeights[cnt] = solver.Constant(weights[cnt]);
}
// contributors to the weight (weight * variable value)
CspTerm[] contributors = new CspTerm[NumItems];
for (int cnt = 0; cnt < NumItems; cnt++)
{
contributors[cnt] = itemWeights[cnt] * itemQty[cnt];
}
// single constraint
CspTerm knapSackCapacity = solver.GreaterEqual(knapsackSize, solver.Sum(contributors));
solver.AddConstraints(knapSackCapacity);
// must be efficient
CspTerm knapSackAtLeast = solver.LessEqual(knapsackSize * efficiency, solver.Sum(contributors));
solver.AddConstraints(knapSackAtLeast);
// start counter and allocate a list for the results.
int nanswers = 0;
patterns = new List <int[]>();
ConstraintSolverSolution sol = solver.Solve();
while (sol.HasFoundSolution)
{
int[] pattern = new int[NumItems];
// extract this pattern from the enumeration.
for (int cnt = 0; cnt < NumItems; cnt++)
{
object val;
sol.TryGetValue(itemQty[cnt], out val);
pattern[cnt] = (int)val;
}
// add it to the output.
patterns.Add(pattern);
nanswers++;
// stop if we reach the limit of results.
if (maxAnswers > 0 && nanswers >= maxAnswers)
{
break;
}
sol.GetNext();
}
}
public override void AddConstaint()
{
ConstraintSystem solver = ConstraintSystemSolver.Instance.Solver;
CspTerm constraint = null;
CspTerm[] inputTerms = new CspTerm[Input.Count];
for (int i = 0; i < Input.Count; i++)
{
inputTerms[i] = Input[i].CspTerm;
}
CspTerm outputTerm = Output.CspTerm;
Type consType = type;
if (IsNotHealthy)
{
// In case the gate is Broken (Not Healthy) - we don't want to add any constraint!!!
return;
/*
* switch (type)
* {
* case Type.and:
* consType = Type.nand;
* break;
* case Type.nand:
* consType = Type.and;
* break;
* case Type.or:
* consType = Type.nor;
* break;
* case Type.nor:
* consType = Type.or;
* break;
* case Type.xor:
* consType = Type.nxor;
* break;
* case Type.nxor:
* consType = Type.xor;
* break;
* }
*/
}
lock (ConstraintSystemSolver.Instance.Locker)
{
//Debug.WriteLine("SAT IN!");
switch (consType)
{
case Type.and:
CspTerm allAndInputs = solver.And(inputTerms);
constraint = solver.Equal(allAndInputs, outputTerm);
break;
case Type.nand:
CspTerm allNandInputs = solver.And(inputTerms);
constraint = solver.Equal(allNandInputs, solver.Not(outputTerm));
break;
case Type.nor:
CspTerm allNorInputs = solver.Or(inputTerms);
constraint = solver.Equal(allNorInputs, solver.Not(outputTerm));
break;
case Type.or:
CspTerm allOrInputs = solver.Or(inputTerms);
constraint = solver.Equal(allOrInputs, outputTerm);
break;
case Type.xor:
//XOR is also:
//http://en.wikipedia.org/wiki/XOR_gate#/media/File:254px_3gate_XOR.jpg
CspTerm firstNand = solver.Not(solver.And(inputTerms));
CspTerm firstOr = solver.Or(inputTerms);
CspTerm secendAnd = solver.And(firstNand, firstOr);
constraint = solver.Equal(secendAnd, outputTerm);
break;
case Type.nxor:
//XOR is also:
//http://en.wikipedia.org/wiki/XOR_gate#/media/File:254px_3gate_XOR.jpg
CspTerm firstNand2 = solver.Not(solver.And(inputTerms));
CspTerm firstOr2 = solver.Or(inputTerms);
CspTerm secendAnd2 = solver.And(firstNand2, firstOr2);
constraint = solver.Equal(secendAnd2, solver.Not(outputTerm));
break;
}
solver.AddConstraints(constraint);
//Debug.WriteLine("SAT OUT!");
}
}
/// <summary>
/// Generates a constraint system based on a variability model. The constraint system can be used to check for satisfiability of configurations as well as optimization.
/// </summary>
/// <param name="variables">Empty input, outputs a list of CSP terms that correspond to the configuration options of the variability model</param>
/// <param name="optionToTerm">A map to get for a given configuration option the corresponding CSP term of the constraint system</param>
/// <param name="termToOption">A map that gives for a given CSP term the corresponding configuration option of the variability model</param>
/// <param name="vm">The variability model for which we generate a constraint system</param>
/// <returns>The generated constraint system consisting of logical terms representing configuration options as well as their boolean constraints.</returns>
internal static ConstraintSystem GetGeneralConstraintSystem(out Dictionary <CspTerm, bool> variables, out Dictionary <ConfigurationOption, CspTerm> optionToTerm, out Dictionary <CspTerm, ConfigurationOption> termToOption, VariabilityModel vm)
{
ConstraintSystem S = ConstraintSystem.CreateSolver();
optionToTerm = new Dictionary <ConfigurationOption, CspTerm>();
termToOption = new Dictionary <CspTerm, ConfigurationOption>();
variables = new Dictionary <CspTerm, bool>();
foreach (ConfigurationOption o in vm.getOptions())
{
CspDomain binDomain = S.DefaultBoolean;
CspTerm temp;
if (o is BinaryOption)
{
temp = S.CreateVariable(binDomain, o);
}
else
{
NumericOption numOpt = (NumericOption)o;
temp = S.CreateVariable(S.CreateIntegerInterval((int)numOpt.Min_value, (int)numOpt.Max_value), o);
}
optionToTerm.Add(o, temp);
termToOption.Add(temp, o);
if (o is NumericOption)
{
variables.Add(temp, false);
}
else
{
variables.Add(temp, true);
}
}
List <List <ConfigurationOption> > alreadyHandledAlternativeOptions = new List <List <ConfigurationOption> >();
//Constraints of a single configuration option
foreach (ConfigurationOption current in vm.getOptions())
{
CspTerm cT = optionToTerm[current];
if (current.Parent == null || current.Parent == vm.Root)
{
if ((current is BinaryOption && ((BinaryOption)current).Optional == false && current.Excluded_Options.Count == 0))
{
S.AddConstraints(S.Implies(S.True, cT));
}
else
{
S.AddConstraints(S.Implies(cT, optionToTerm[vm.Root]));
}
}
if (current.Parent != null && current.Parent != vm.Root)
{
CspTerm parent = optionToTerm[(BinaryOption)current.Parent];
S.AddConstraints(S.Implies(cT, parent));
if (current is BinaryOption && ((BinaryOption)current).Optional == false && current.Excluded_Options.Count == 0)
{
S.AddConstraints(S.Implies(parent, cT));//mandatory child relationship
}
}
// Add numeric integer values
if (current is NumericOption)
{
NumericOption numOpt = (NumericOption)current;
List <double> values = numOpt.getAllValues();
List <CspTerm> equals = new List <CspTerm>();
foreach (double d in values)
{
equals.Add(S.Equal((int)d, cT));
}
S.AddConstraints(S.Or(equals.ToArray()));
}
//Alternative or other exclusion constraints
if (current.Excluded_Options.Count > 0 && current is BinaryOption)
{
BinaryOption binOpt = (BinaryOption)current;
List <ConfigurationOption> alternativeOptions = binOpt.collectAlternativeOptions();
if (alternativeOptions.Count > 0)
{
//Check whether we handled this group of alternatives already
foreach (var alternativeGroup in alreadyHandledAlternativeOptions)
{
foreach (var alternative in alternativeGroup)
{
if (current == alternative)
{
goto handledAlternative;
}
}
}
//It is not allowed that an alternative group has no parent element
CspTerm parent = null;
if (current.Parent == null)
{
parent = S.True;
}
else
{
parent = optionToTerm[(BinaryOption)current.Parent];
}
CspTerm[] terms = new CspTerm[alternativeOptions.Count + 1];
terms[0] = cT;
int i = 1;
foreach (BinaryOption altEle in alternativeOptions)
{
CspTerm temp = optionToTerm[altEle];
terms[i] = temp;
i++;
}
S.AddConstraints(S.Implies(parent, S.ExactlyMofN(1, terms)));
alreadyHandledAlternativeOptions.Add(alternativeOptions);
handledAlternative : { }
}
//Excluded option(s) as cross-tree constraint(s)
List <List <ConfigurationOption> > nonAlternative = binOpt.getNonAlternativeExlcudedOptions();
if (nonAlternative.Count > 0)
{
foreach (var excludedOption in nonAlternative)
{
CspTerm[] orTerm = new CspTerm[excludedOption.Count];
int i = 0;
foreach (var opt in excludedOption)
{
CspTerm target = optionToTerm[(BinaryOption)opt];
orTerm[i] = target;
i++;
}
S.AddConstraints(S.Implies(cT, S.Not(S.Or(orTerm))));
}
}
}
//Handle implies
if (current.Implied_Options.Count > 0)
{
foreach (List <ConfigurationOption> impliedOr in current.Implied_Options)
{
CspTerm[] orTerms = new CspTerm[impliedOr.Count];
//Possible error: if a binary option impies a numeric option
for (int i = 0; i < impliedOr.Count; i++)
{
orTerms[i] = optionToTerm[(BinaryOption)impliedOr.ElementAt(i)];
}
S.AddConstraints(S.Implies(optionToTerm[current], S.Or(orTerms)));
}
}
}
//Handle global cross-tree constraints involving multiple options at a time
// the constraints should be in conjunctive normal form
foreach (string constraint in vm.BinaryConstraints)
{
bool and = false;
string[] terms;
if (constraint.Contains("&"))
{
and = true;
terms = constraint.Split('&');
}
else
{
terms = constraint.Split('|');
}
CspTerm[] cspTerms = new CspTerm[terms.Count()];
int i = 0;
foreach (string t in terms)
{
string optName = t.Trim();
if (optName.StartsWith("-") || optName.StartsWith("!"))
{
optName = optName.Substring(1);
BinaryOption binOpt = vm.getBinaryOption(optName);
CspTerm cspElem = optionToTerm[binOpt];
CspTerm notCspElem = S.Not(cspElem);
cspTerms[i] = notCspElem;
}
else
{
BinaryOption binOpt = vm.getBinaryOption(optName);
CspTerm cspElem = optionToTerm[binOpt];
cspTerms[i] = cspElem;
}
i++;
}
if (and)
{
S.AddConstraints(S.And(cspTerms));
}
else
{
S.AddConstraints(S.Or(cspTerms));
}
}
return(S);
}
/// <summary>
/// Generates a constraint system based on a variability model. The constraint system can be used to check for satisfiability of configurations as well as optimization.
/// </summary>
/// <param name="variables">Empty input, outputs a list of CSP terms that correspond to the configuration options of the variability model</param>
/// <param name="optionToTerm">A map to get for a given configuration option the corresponding CSP term of the constraint system</param>
/// <param name="termToOption">A map that gives for a given CSP term the corresponding configuration option of the variability model</param>
/// <param name="vm">The variability model for which we generate a constraint system</param>
/// <returns>The generated constraint system consisting of logical terms representing configuration options as well as their boolean constraints.</returns>
internal static ConstraintSystem getConstraintSystem(out List <CspTerm> variables, out Dictionary <BinaryOption, CspTerm> optionToTerm, out Dictionary <CspTerm, BinaryOption> termToOption, VariabilityModel vm)
{
//Reusing seems to not work correctely. The problem: configurations are realized as additional constraints for the system.
//however, when checking for the next config, the old config's constraints remain in the solver such that we have a wrong result.
/*
* if (csystem != null && variables_global != null && optionToTerm_global != null && termToOption_global != null && vm != null)
* {//For optimization purpose
* if (vm.BinaryOptions.Count == vm_global.BinaryOptions.Count && vm.Name.Equals(vm_global.Name))
* {
* variables = variables_global;
* optionToTerm = optionToTerm_global;
* termToOption = termToOption_global;
* return csystem;
* }
* }*/
ConstraintSystem S = ConstraintSystem.CreateSolver();
optionToTerm = new Dictionary <BinaryOption, CspTerm>();
termToOption = new Dictionary <CspTerm, BinaryOption>();
variables = new List <CspTerm>();
foreach (BinaryOption binOpt in vm.BinaryOptions)
{
CspDomain domain = S.DefaultBoolean;
CspTerm temp = S.CreateVariable(domain, binOpt);
optionToTerm.Add(binOpt, temp);
termToOption.Add(temp, binOpt);
variables.Add(temp);
}
List <List <ConfigurationOption> > alreadyHandledAlternativeOptions = new List <List <ConfigurationOption> >();
//Constraints of a single configuration option
foreach (BinaryOption current in vm.BinaryOptions)
{
CspTerm cT = optionToTerm[current];
if (current.Parent == null || current.Parent == vm.Root)
{
if (current.Optional == false && current.Excluded_Options.Count == 0)
{
S.AddConstraints(S.Implies(S.True, cT));
}
else
{
S.AddConstraints(S.Implies(cT, optionToTerm[vm.Root]));
}
}
if (current.Parent != null && current.Parent != vm.Root)
{
CspTerm parent = optionToTerm[(BinaryOption)current.Parent];
S.AddConstraints(S.Implies(cT, parent));
if (current.Optional == false && current.Excluded_Options.Count == 0)
{
S.AddConstraints(S.Implies(parent, cT));//mandatory child relationship
}
}
//Alternative or other exclusion constraints
if (current.Excluded_Options.Count > 0)
{
List <ConfigurationOption> alternativeOptions = current.collectAlternativeOptions();
if (alternativeOptions.Count > 0)
{
//Check whether we handled this group of alternatives already
foreach (var alternativeGroup in alreadyHandledAlternativeOptions)
{
foreach (var alternative in alternativeGroup)
{
if (current == alternative)
{
goto handledAlternative;
}
}
}
//It is not allowed that an alternative group has no parent element
CspTerm parent = null;
if (current.Parent == null)
{
parent = S.True;
}
else
{
parent = optionToTerm[(BinaryOption)current.Parent];
}
CspTerm[] terms = new CspTerm[alternativeOptions.Count + 1];
terms[0] = cT;
int i = 1;
foreach (BinaryOption altEle in alternativeOptions)
{
CspTerm temp = optionToTerm[altEle];
terms[i] = temp;
i++;
}
S.AddConstraints(S.Implies(parent, S.ExactlyMofN(1, terms)));
alreadyHandledAlternativeOptions.Add(alternativeOptions);
handledAlternative : { }
}
//Excluded option(s) as cross-tree constraint(s)
List <List <ConfigurationOption> > nonAlternative = current.getNonAlternativeExlcudedOptions();
if (nonAlternative.Count > 0)
{
foreach (var excludedOption in nonAlternative)
{
CspTerm[] orTerm = new CspTerm[excludedOption.Count];
int i = 0;
foreach (var opt in excludedOption)
{
CspTerm target = optionToTerm[(BinaryOption)opt];
orTerm[i] = target;
i++;
}
S.AddConstraints(S.Implies(cT, S.Not(S.Or(orTerm))));
}
}
}
//Handle implies
if (current.Implied_Options.Count > 0)
{
foreach (List <ConfigurationOption> impliedOr in current.Implied_Options)
{
CspTerm[] orTerms = new CspTerm[impliedOr.Count];
//Possible error: if a binary option impies a numeric option
for (int i = 0; i < impliedOr.Count; i++)
{
orTerms[i] = optionToTerm[(BinaryOption)impliedOr.ElementAt(i)];
}
S.AddConstraints(S.Implies(optionToTerm[current], S.Or(orTerms)));
}
}
}
//Handle global cross-tree constraints involving multiple options at a time
// the constraints should be in conjunctive normal form
foreach (string constraint in vm.BinaryConstraints)
{
bool and = false;
string[] terms;
if (constraint.Contains("&"))
{
and = true;
terms = constraint.Split('&');
}
else
{
terms = constraint.Split('|');
}
CspTerm[] cspTerms = new CspTerm[terms.Count()];
int i = 0;
foreach (string t in terms)
{
string optName = t.Trim();
if (optName.StartsWith("-") || optName.StartsWith("!"))
{
optName = optName.Substring(1);
BinaryOption binOpt = vm.getBinaryOption(optName);
CspTerm cspElem = optionToTerm[binOpt];
CspTerm notCspElem = S.Not(cspElem);
cspTerms[i] = notCspElem;
}
else
{
BinaryOption binOpt = vm.getBinaryOption(optName);
CspTerm cspElem = optionToTerm[binOpt];
cspTerms[i] = cspElem;
}
i++;
}
if (and)
{
S.AddConstraints(S.And(cspTerms));
}
else
{
S.AddConstraints(S.Or(cspTerms));
}
}
csystem = S;
optionToTerm_global = optionToTerm;
vm_global = vm;
termToOption_global = termToOption;
variables_global = variables;
return(S);
}
static void Main(string[] args)
{
try
{
Console.WriteLine("Seleccione el método por el que desea resolver el problema:\n1 Programación por restricciones\n2 Programación Lineal");
switch (int.Parse(Console.ReadLine()))
{
case 1:
SolverContext context = new SolverContext();
Model model = context.CreateModel();
// Creación de variables
Decision x1 = new Decision(Domain.Integer, "x1");
model.AddDecision(x1);
Decision x2 = new Decision(Domain.Integer, "x2");
model.AddDecision(x2);
Decision x3 = new Decision(Domain.Integer, "x3");
model.AddDecision(x3);
// Creación de limites
model.AddConstraint("limitX1", 50 <= x1 <= 300);
model.AddConstraint("limitX2", 100 <= x2 <= 200);
model.AddConstraint("limitX3", 20 <= x3 <= 1000);
// Creación de restricciones
model.AddConstraint("restriccion1", 200 <= (x1 + x2 + x3) <= 280);
model.AddConstraint("restriccion2", 100 <= (x1 + (3 * x3)) <= 2000);
model.AddConstraint("restriccion3", 50 <= ((2 + x1) + (4 * x3)) <= 1000);
// Función objetivo
model.AddGoal("maximo", GoalKind.Maximize, -(4 * x1) - (2 * x2) + x3);
// Solucion
Solution solucion = context.Solve(new SimplexDirective());
Report reporte = solucion.GetReport();
// Imprimir
Console.WriteLine(reporte);
Console.ReadLine();
break;
case 2:
//Solucionador específico
ConstraintSystem csp = ConstraintSystem.CreateSolver();
// Creacíón de variables
CspTerm sx1 = csp.CreateVariable(csp.CreateIntegerInterval(50, 300), "x1");
CspTerm sx2 = csp.CreateVariable(csp.CreateIntegerInterval(100, 200), "x2");
CspTerm sx3 = csp.CreateVariable(csp.CreateIntegerInterval(20, 1000), "x3");
// Creación de restricciones
csp.AddConstraints(200 <= (sx1 + sx2 + sx3) <= 280,
100 <= sx1 + (3 * sx2) <= 2000,
50 <= (2 * sx1) + (4 * sx3) <= 1000);
// Solución
ConstraintSolverSolution cspSolucion = csp.Solve();
int numero = 1;
while (cspSolucion.HasFoundSolution)
{
object rx1, rx2, rx3;
if (!cspSolucion.TryGetValue(sx1, out rx1) || !cspSolucion.TryGetValue(sx2, out rx2) || !cspSolucion.TryGetValue(sx3, out rx3))
{
throw new InvalidProgramException("No se encontro solución");
}
Console.WriteLine(String.Format("Solución {0} :\nx1={1}\nx2={2}\nx3={3}", numero, rx1, rx2, rx3));
numero += 1;
cspSolucion.GetNext();
}
/*
* //Solucionador específico
* SimplexSolver sSolver = new SimplexSolver();
* //Creación de variables
* int sx1, sx2, sx3;
* sSolver.AddVariable("x1", out sx1);
* sSolver.SetBounds(sx1, 50, 300);
* sSolver.AddVariable("x2", out sx2);
* sSolver.SetBounds(sx2, 100, 200);
* sSolver.AddVariable("x3", out sx3);
* sSolver.SetBounds(sx3,20,1000);
* //Creación de restricciones
* int r1, r2, r3, goal;
* sSolver.AddRow("restriccion1", out r1);
* sSolver.SetCoefficient(r1, sx1, 1);
* sSolver.SetCoefficient(r1, sx2, 1);
* sSolver.SetCoefficient(r1, sx3, 1);
* sSolver.SetBounds(r1, 200, 280);
* sSolver.AddRow("restriccion2", out r2);
* sSolver.SetCoefficient(r2, sx1, 1);
* sSolver.SetCoefficient(r2, sx2, 3);
* sSolver.SetBounds(r2, 100, 2000);
* sSolver.AddRow("restriccion3", out r3);
* sSolver.SetCoefficient(r3, sx1, 2);
* sSolver.SetCoefficient(r3, sx3, 4);
* sSolver.SetBounds(r3, 50, 1000);
* //Función objetivo
* sSolver.AddRow("objetivo", out goal);
* sSolver.SetCoefficient(goal, sx1, -4);
* sSolver.SetCoefficient(goal, sx2, -2);
* sSolver.SetCoefficient(goal, sx3, 1);
* sSolver.SetBounds(goal, Rational.NegativeInfinity,Rational.PositiveInfinity);
* sSolver.AddGoal(goal, 1, false);
* //Solución
* sSolver.Solve(new SimplexSolverParams());
* sSolver.GetReport();
*/
break;
}
}
catch (Exception ex)
{
Console.WriteLine(ex.Message);
Console.WriteLine(ex.StackTrace);
Console.ReadLine();
}
}
///// <summary> Helper function for reading the Bus Driver data file
///// </summary>
//static List<string> Numerals(string line)
//{
// List<string> result = new List<string>();
// int left = 0;
// while (left < line.Length)
// {
// char c = line[left];
// if (('0' <= c) && (c <= '9'))
// {
// int right = left + 1;
// while ((right < line.Length) && ('0' <= line[right]) && (line[right] <= '9'))
// right++;
// result.Add(line.Substring(left, right - left));
// left = right + 1;
// }
// else
// left++;
// }
// return result;
//}
/// <summary> Bus Drivers. Data taken from data files of London bus companies, with the
/// problem being to find the cheapest, complete, non-overlapping set of task
/// assignments that will give a feasible schedule.
/// </summary>
public static void BusDrivers(string sourceFilePath)
{
// http://www-old.cs.st-andrews.ac.uk/~ianm/CSPLib/prob/prob022/index.html
ConstraintSystem S = ConstraintSystem.CreateSolver();
List <CspTerm> driverCosts = new List <CspTerm>();
List <int[]> driversTasks = new List <int[]>();
int nTasks = 0;
// Read the data file. Each row specifies a driver cost, a count of tasks, and the task numbers
try
{
using (StreamReader sr = new StreamReader(sourceFilePath))
{
String line;
while ((line = sr.ReadLine()) != null)
{
int[] tasks;
driverCosts.Add(S.Constant(ReadDriver(line, out tasks)));
nTasks += tasks.Length;
Array.Sort <int>(tasks);
driversTasks.Add(tasks);
}
}
int nDrivers = driversTasks.Count;
// create a master list of tasks by sorting the raw union and then compressing out duplicates.
int[] allTasks = new int[nTasks];
nTasks = 0;
foreach (int[] tasks in driversTasks)
{
foreach (int x in tasks)
{
allTasks[nTasks++] = x;
}
}
Array.Sort <int>(allTasks);
nTasks = 0;
for (int i = 1; i < allTasks.Length; i++)
{
if (allTasks[nTasks] < allTasks[i])
{
allTasks[++nTasks] = allTasks[i];
}
}
nTasks++;
Array.Resize <int>(ref allTasks, nTasks);
// We now have an array of all the tasks, and a list of all the drivers.
// The problem statement comes down to:
// - each task must be assigned exactly once
// - minimize the cost of drivers
// We add a boolean vector representing the drivers, true if the driver is to be used.
CspTerm[] driversUsed = S.CreateBooleanVector("drivers", nDrivers); // these are the Decision Variables
// We now create an array which maps which tasks are in which drivers.
// In addition to this static map, we create a dynamic map of the usage and the costs.
CspTerm[][] taskActualUse = new CspTerm[nTasks][];
CspTerm[] driverActualCost = new CspTerm[nDrivers];
for (int t = 0; t < nTasks; t++)
{
taskActualUse[t] = new CspTerm[nDrivers];
for (int r = 0; r < nDrivers; r++)
{
taskActualUse[t][r] = (0 <= Array.BinarySearch <int>(driversTasks[r], allTasks[t])) ? driversUsed[r] : S.False;
}
S.AddConstraints(
S.ExactlyMofN(1, taskActualUse[t]) // this task appears exactly once
);
}
// set the goal
for (int r = 0; r < nDrivers; r++)
{
driverActualCost[r] = driversUsed[r] * driverCosts[r]; // dynamic cost map
}
S.TryAddMinimizationGoals(S.Sum(driverActualCost));
// now run the Solver and print the solutions
int solnId = 0;
ConstraintSolverSolution soln = S.Solve();
if (soln.HasFoundSolution)
{
System.Console.WriteLine("Solution #" + solnId++);
for (int d = 0; d < driversUsed.Length; d++)
{
object isUsed;
if (!soln.TryGetValue(driversUsed[d], out isUsed))
{
throw new InvalidProgramException("can't find drive in the solution: " + d.ToString());
}
// Take only the decision variables which turn out true.
// For each true row, print the row number and the list of tasks.
if (1 == (int)isUsed)
{
StringBuilder line = new StringBuilder(d.ToString());
line.Append(": ");
foreach (int x in driversTasks[d])
{
line.Append(x.ToString()).Append(", ");
}
System.Console.WriteLine(line.ToString());
}
}
}
if (solnId == 0)
{
System.Console.WriteLine("No solution found.");
}
}
catch (Exception e)
{
// Let the user know what went wrong.
Console.WriteLine("The file could not be read:");
Console.WriteLine(e.Message);
}
}
/// <summary>
/// Based on a given (partial) configuration and a variability, we aim at finding all optimally maximal or minimal (in terms of selected binary options) configurations.
/// </summary>
/// <param name="config">The (partial) configuration which needs to be expaned to be valid.</param>
/// <param name="vm">Variability model containing all options and their constraints.</param>
/// <param name="minimize">If true, we search for the smallest (in terms of selected options) valid configuration. If false, we search for the largest one.</param>
/// <param name="unwantedOptions">Binary options that we do not want to become part of the configuration. Might be part if there is no other valid configuration without them</param>
/// <returns>A list of configurations that satisfies the VM and the goal (or null if there is none).</returns>
public List <List <BinaryOption> > MaximizeConfig(List <BinaryOption> config, VariabilityModel vm, bool minimize, List <BinaryOption> unwantedOptions)
{
List <CspTerm> variables = new List <CspTerm>();
Dictionary <BinaryOption, CspTerm> elemToTerm = new Dictionary <BinaryOption, CspTerm>();
Dictionary <CspTerm, BinaryOption> termToElem = new Dictionary <CspTerm, BinaryOption>();
ConstraintSystem S = CSPsolver.getConstraintSystem(out variables, out elemToTerm, out termToElem, vm);
//Feature Selection
if (config != null)
{
foreach (BinaryOption binOpt in config)
{
CspTerm term = elemToTerm[binOpt];
S.AddConstraints(S.Implies(S.True, term));
}
}
//Defining Goals
CspTerm[] finalGoals = new CspTerm[variables.Count];
for (int r = 0; r < variables.Count; r++)
{
if (minimize == true)
{
BinaryOption binOpt = termToElem[variables[r]];
if (unwantedOptions != null && (unwantedOptions.Contains(binOpt) && !config.Contains(binOpt)))
{
finalGoals[r] = variables[r] * 10000;
}
else
{
// Element is part of an altnerative Group ... we want to select always the same option of the group, so we give different weights to the member of the group
//Functionality deactivated... todo needs further handling
/*if (binOpt.getAlternatives().Count != 0)
* {
* finalGoals[r] = variables[r] * (binOpt.getID() * 10);
* }
* else
* {*/
finalGoals[r] = variables[r] * 1;
//}
// wenn in einer alternative, dann bekommt es einen wert nach seiner reihenfolge
// id mal 10
}
}
else
{
finalGoals[r] = variables[r] * -1; // dynamic cost map
}
}
S.TryAddMinimizationGoals(S.Sum(finalGoals));
ConstraintSolverSolution soln = S.Solve();
List <string> erg2 = new List <string>();
List <BinaryOption> tempConfig = new List <BinaryOption>();
List <List <BinaryOption> > resultConfigs = new List <List <BinaryOption> >();
while (soln.HasFoundSolution && soln.Quality == ConstraintSolverSolution.SolutionQuality.Optimal)
{
tempConfig.Clear();
foreach (CspTerm cT in variables)
{
if (soln.GetIntegerValue(cT) == 1)
{
tempConfig.Add(termToElem[cT]);
}
}
if (minimize && tempConfig != null)
{
resultConfigs.Add(tempConfig);
break;
}
if (!Configuration.containsBinaryConfiguration(resultConfigs, tempConfig))
{
resultConfigs.Add(tempConfig);
}
soln.GetNext();
}
return(resultConfigs);
}
private static ConstraintSystem AddConstraintsToSolver(
ConstraintSystem solver, ZebraPuzzleConstraints constraints)
{
solver.AddConstraints(
constraints.TheEnglishManLivesInTheRedHouse());
solver.AddConstraints(
constraints.TheSwedeHasADog());
solver.AddConstraints(
constraints.TheDaneDrinksTea());
solver.AddConstraints(
constraints.TheGreenHouseIsImmediatelyLeftOfTheWhiteHouse());
solver.AddConstraints(
constraints.TheyDrinkCoffeeInTheGreenHouse());
solver.AddConstraints(
constraints.TheManWhoSmokesPallMallHasBirds());
solver.AddConstraints(
constraints.InTheYellowHouseTheySmokeDunhill());
solver.AddConstraints(
constraints.InTheMiddleHouseTheyDrinkMilk());
solver.AddConstraints(
constraints.TheNorwegianLivesInTheFirstHouse());
solver.AddConstraints(
constraints.TheManWhoSmokesBlendInHouseNextToHouseWithCats());
solver.AddConstraints(
constraints.InHouseNextToHouseWhereHaveAHorseSmokeDunhill());
solver.AddConstraints(
constraints.TheManWhoSmokesBlueMasterDrinksBeer());
solver.AddConstraints(
constraints.TheGermanSmokesPrince());
solver.AddConstraints(
constraints.TheNorwegianLivesNextToTheBlueHouse());
solver.AddConstraints(
constraints.TheyDrinkWaterInHouseNextToHouseWhereSmokeBlend());
return(solver);
}
public ConstraintSolverSolution SolveB(int maxSolutions = 100)
{
ConstraintSystem S = ConstraintSystem.CreateSolver();
//Define how many agents may work per shift
CspDomain cspShiftsForceDomain = S.CreateIntegerInterval(first: 0, last: MaxAgents);
//Decision variables, one per shift, that will hold the result of how may agents must work per shift to fullfil requirements
CspTerm[] cspShiftsForce = S.CreateVariableVector(domain: cspShiftsForceDomain, key: "force", length: Shifts.Count);
//index shifts, their variable CspTerm by the shift's relative no (0=first, 1=second, etc)
ShiftsX = new Dictionary <Models.Shift, CspTerm>();
int i = 0;
Shifts.ForEach(x => { ShiftsX.Add(x, cspShiftsForce[i]); i++; });
//Constraint - Agents on every half hour must be >= requirement for that half hour
List <CspTerm> cspHalfHourExcess = new List <CspTerm>();
foreach (var halfHourRq in HalfHourRequirements)
{
//find shifts including that halftime, and calculate their sum of force
List <CspTerm> cspActive = new List <CspTerm>();
foreach (var entry in ShiftsX)
{
if (entry.Key.IncludesHalfHour(halfHourRq.Start))
{
cspActive.Add(entry.Value);
}
}
//add constraint for sum of force of active shifts on that halfhour
//if we need agents but no shifts exists for a halfhour, do not add a constraint
if (cspActive.Count > 0)
{
var s = S.Sum(cspActive.ToArray()) - S.Constant(halfHourRq.RequiredForce);
S.AddConstraints(
S.LessEqual(constant: 0, inputs: s)
);
cspHalfHourExcess.Add(s);
}
}
//var goal = S.Sum(shiftsX.Values.ToArray());
//bool ok = S.TryAddMinimizationGoals(goal);
var goalMinExcess = S.Sum(cspHalfHourExcess.ToArray());
bool ok = S.TryAddMinimizationGoals(goalMinExcess);
ConstraintSolverSolution solution = S.Solve();
Console.WriteLine();
GetSolutionsAll(solution: solution, maxSolutions: maxSolutions);
if (ShiftsForce == null || ShiftsForce.Count == 0)
{
Console.WriteLine("No solution found");
}
if (ShiftsForce != null)
{
foreach (var shiftForceEntry in ShiftsForce)
{
ShowSolution(no: shiftForceEntry.Key, shiftsForce: shiftForceEntry.Value, showAgents: true, showSlots: false);
}
}
return(solution);
}
public ConstraintSystem PrepareCspSolver()
{
ConstraintSystem S = ConstraintSystem.CreateSolver();
//Define how many agents may work per shift
var maxRq = HalfHourRequirements.Max(x => x.RequiredForce);
CspDomain cspShiftsForceDomain = S.CreateIntegerInterval(first: 0, last: maxRq);
//var cspShiftsForceDomain = S.CreateIntegerSet(new int[] { 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40 });
//Decision variables, one per shift, that will hold the result of how may agents must work per shift to fullfil requirements
CspTerm[] cspShiftsForce = S.CreateVariableVector(domain: cspShiftsForceDomain, key: "force", length: Shifts.Count);
//index shifts, their variable CspTerm by the shift's relative no (0=first, 1=second, etc)
ShiftsX = new Dictionary <Models.Shift, CspTerm>();
int i = 0;
Shifts.ForEach(x => { ShiftsX.Add(x, cspShiftsForce[i]); i++; });
//Constraint - Agents from every active shift on every half hour must be >= requirement for that half hour
List <CspTerm> cspHalfHourExcess = new List <CspTerm>();
foreach (var halfHourRq in HalfHourRequirements)
{
//find shifts including that halftime, and calculate their sum of force
List <CspTerm> cspActive = new List <CspTerm>();
foreach (var entry in ShiftsX)
{
if (entry.Key.IncludesHalfHour(halfHourRq.Start))
{
cspActive.Add(entry.Value);
}
}
//add constraint for sum of force of active shifts on that halfhour
//if we need agents but no shifts exists for a halfhour, do not add a constraint
if (cspActive.Count > 0)
{
//var s = S.Sum(cspActive.ToArray()) - S.Constant(halfHourRq.RequiredForce);
S.AddConstraints(
S.LessEqual(constant: halfHourRq.RequiredForce, inputs: S.Sum(cspActive.ToArray()))
);
//cspHalfHourExcess.Add(s);
}
}
//if (false && cspHalfHourExcess.Count > 0)
// S.AddConstraints(
// S.LessEqual(constant: 0, inputs: S.Sum(cspHalfHourExcess.ToArray()))
// );
bool xx = true;
if (xx)
{
var goal = S.Sum(ShiftsX.Values.ToArray());
bool ok = S.TryAddMinimizationGoals(goal);
}
else
{
//S.RemoveAllMinimizationGoals();
var goalMinExcess = S.Sum(cspHalfHourExcess.ToArray());
bool ok = S.TryAddMinimizationGoals(goalMinExcess);
}
return(S);
}
/// <summary>
/// This method searches for a corresponding methods in the dynamically loaded assemblies and calls it if found. It prefers due to performance reasons the Microsoft Solver Foundation implementation.
/// </summary>
/// <param name="config">The (partial) configuration which needs to be expaned to be valid.</param>
/// <param name="vm">Variability model containing all options and their constraints.</param>
/// <param name="minimize">If true, we search for the smallest (in terms of selected options) valid configuration. If false, we search for the largest one.</param>
/// <param name="unWantedOptions">Binary options that we do not want to become part of the configuration. Might be part if there is no other valid configuration without them.</param>
/// <returns>The valid configuration (or null if there is none) that satisfies the VM and the goal.</returns>
public List <BinaryOption> MinimizeConfig(List <BinaryOption> config, VariabilityModel vm, bool minimize, List <BinaryOption> unWantedOptions)
{
List <CspTerm> variables = new List <CspTerm>();
Dictionary <BinaryOption, CspTerm> elemToTerm = new Dictionary <BinaryOption, CspTerm>();
Dictionary <CspTerm, BinaryOption> termToElem = new Dictionary <CspTerm, BinaryOption>();
ConstraintSystem S = CSPsolver.getConstraintSystem(out variables, out elemToTerm, out termToElem, vm);
//Feature Selection
foreach (BinaryOption binOpt in config)
{
CspTerm term = elemToTerm[binOpt];
S.AddConstraints(S.Implies(S.True, term));
}
//Defining Goals
CspTerm[] finalGoals = new CspTerm[variables.Count];
for (int r = 0; r < variables.Count; r++)
{
if (minimize == true)
{
if (unWantedOptions != null && (unWantedOptions.Contains(termToElem[variables[r]]) && !config.Contains(termToElem[variables[r]])))
{
finalGoals[r] = variables[r] * 100;
}
else
{
finalGoals[r] = variables[r] * 1;
}
}
else
{
finalGoals[r] = variables[r] * -1; // dynamic cost map
}
}
S.TryAddMinimizationGoals(S.Sum(finalGoals));
ConstraintSolverSolution soln = S.Solve();
List <string> erg2 = new List <string>();
List <BinaryOption> tempConfig = new List <BinaryOption>();
while (soln.HasFoundSolution)
{
tempConfig.Clear();
foreach (CspTerm cT in variables)
{
if (soln.GetIntegerValue(cT) == 1)
{
tempConfig.Add(termToElem[cT]);
}
}
if (minimize && tempConfig != null)
{
break;
}
soln.GetNext();
}
return(tempConfig);
}
public static void Run()
{
ConstraintSystem S = ConstraintSystem.CreateSolver();
List <KeyValuePair <CspTerm, string> > termList = new List <KeyValuePair <CspTerm, string> >();
// create a Term between [1..5], associate it with a name for later ease of display
NamedTerm namedTerm = delegate(string name) {
CspTerm x = S.CreateVariable(S.CreateIntegerInterval(1, 5), name);
termList.Add(new KeyValuePair <CspTerm, string>(x, name));
return(x);
};
// the people and attributes will all be matched via the house they reside in.
CspTerm English = namedTerm("English"), Spanish = namedTerm("Spanish"), Japanese = namedTerm("Japanese"), Italian = namedTerm("Italian"), Norwegian = namedTerm("Norwegian");
CspTerm red = namedTerm("red"), green = namedTerm("green"), white = namedTerm("white"), blue = namedTerm("blue"), yellow = namedTerm("yellow");
CspTerm dog = namedTerm("dog"), snails = namedTerm("snails"), fox = namedTerm("fox"), horse = namedTerm("horse"), zebra = namedTerm("zebra");
CspTerm painter = namedTerm("painter"), sculptor = namedTerm("sculptor"), diplomat = namedTerm("diplomat"), violinist = namedTerm("violinist"), doctor = namedTerm("doctor");
CspTerm tea = namedTerm("tea"), coffee = namedTerm("coffee"), milk = namedTerm("milk"), juice = namedTerm("juice"), water = namedTerm("water");
S.AddConstraints(
S.Unequal(English, Spanish, Japanese, Italian, Norwegian),
S.Unequal(red, green, white, blue, yellow),
S.Unequal(dog, snails, fox, horse, zebra),
S.Unequal(painter, sculptor, diplomat, violinist, doctor),
S.Unequal(tea, coffee, milk, juice, water),
S.Equal(English, red),
S.Equal(Spanish, dog),
S.Equal(Japanese, painter),
S.Equal(Italian, tea),
S.Equal(1, Norwegian),
S.Equal(green, coffee),
S.Equal(1, green - white),
S.Equal(sculptor, snails),
S.Equal(diplomat, yellow),
S.Equal(3, milk),
S.Equal(1, S.Abs(Norwegian - blue)),
S.Equal(violinist, juice),
S.Equal(1, S.Abs(fox - doctor)),
S.Equal(1, S.Abs(horse - diplomat))
);
bool unsolved = true;
ConstraintSolverSolution soln = S.Solve();
while (soln.HasFoundSolution)
{
unsolved = false;
System.Console.WriteLine("solved.");
StringBuilder[] houses = new StringBuilder[5];
for (int i = 0; i < 5; i++)
{
houses[i] = new StringBuilder(i.ToString());
}
foreach (KeyValuePair <CspTerm, string> kvp in termList)
{
string item = kvp.Value;
object house;
if (!soln.TryGetValue(kvp.Key, out house))
{
throw new InvalidProgramException("can't find a Term in the solution: " + item);
}
houses[(int)house - 1].Append(", ");
houses[(int)house - 1].Append(item);
}
foreach (StringBuilder house in houses)
{
System.Console.WriteLine(house);
}
soln.GetNext();
}
if (unsolved)
{
System.Console.WriteLine("No solution found.");
}
else
{
System.Console.WriteLine("Solution should have the Norwegian drinking water and the Japanese with the zebra.");
}
}
public static void Run(int teamsNo)
{
// schedule N teams to play N-1 matches (one against every other team) with a difference
// of no more than 1 extra game away or home. Note that N must be even (since every team
// must be paired every week).
ConstraintSystem S = ConstraintSystem.CreateSolver();
// The teams are numbered 0 to N-1 for simplicity in index lookups,
// since our arrays are zero-based.
CspDomain Teams = S.CreateIntegerInterval(0, teamsNo - 1);
CspTerm[][] matches = S.CreateVariableArray(Teams, "opponents", teamsNo, teamsNo - 1);
CspTerm[][] atHome = S.CreateBooleanArray("atHome", teamsNo, teamsNo - 1);
// each row represents the N-1 games the teams play. The 0th week has an even-odd
// assignment since by symmetry that is equivalent to any other assignment and
// we thereby eliminate redundant solutions being enumerated.
for (int t = 0; t < teamsNo; t++)
{
CspTerm atHomeSum = S.Sum(atHome[t]);
S.AddConstraints(
S.Unequal(t, matches[t]), // don't play self, and play every other team
S.LessEqual(teamsNo / 2 - 1, atHomeSum, S.Constant(teamsNo / 2)), // a balance of atHomes
S.Equal(t ^ 1, matches[t][0]) // even-odd pairing in the initial round
);
}
for (int w = 0; w < teamsNo - 1; w++)
{
S.AddConstraints(
S.Unequal(GetColumn(matches, w)) // every team appears once each week
);
for (int t = 0; t < teamsNo; t++)
{
S.AddConstraints(
S.Equal(t, S.Index(matches, matches[t][w], w)), // Each team's pair's pair must be itself.
S.Equal(atHome[t][w], !S.Index(atHome, matches[t][w], w)) // Each pair is Home-Away or Away-Home.
);
}
}
// That's it! The problem is delivered to the solver.
// Now to get an answer...
//bool unsolved = true;
ConstraintSolverSolution soln = S.Solve();
if (soln.HasFoundSolution)
{
//unsolved = false;
Console.Write(" | ");
for (int w = 0; w < teamsNo - 1; w++)
{
Console.Write("{1}Wk{0,2}", w + 1, w == 0 ? "" : " | ");
}
Console.WriteLine();
Console.Write(" | ");
for (int w = 0; w < teamsNo - 1; w++)
{
Console.Write("{1}OP H", w + 1, w == 0 ? "" : " | ");
}
Console.WriteLine();
Console.WriteLine(" {0}", "|" + new String('-', teamsNo * 6));
for (int t = 0; t < teamsNo; t++)
{
StringBuilder line = new StringBuilder();
line.AppendFormat("Team {0,2}| ", t + 1);
for (int w = 0; w < teamsNo - 1; w++)
{
object opponent, home;
if (!soln.TryGetValue(matches[t][w], out opponent))
{
throw new InvalidProgramException(matches[t][w].Key.ToString());
}
if (!soln.TryGetValue(atHome[t][w], out home))
{
throw new InvalidProgramException(atHome[t][w].Key.ToString());
}
line.AppendFormat("{2}{0,2} {1}",
((int)opponent) + 1,
(int)home == 1 ? "*" : " ",
w == 0 ? "" : " | "
);
//line.Append(opponent.ToString());
//line.Append(((int)home == 1) ? " H," : " A,");
}
System.Console.WriteLine(line.ToString());
}
System.Console.WriteLine();
}
else
{
System.Console.WriteLine("No solution found.");
}
}
public static void Run(string parameters)
{
// http://www-old.cs.st-andrews.ac.uk/~ianm/CSPLib/prob/prob022/index.html
ConstraintSystem S = ConstraintSystem.CreateSolver();
List <CspTerm> driverCosts = new List <CspTerm>();
List <int[]> driversTasks = new List <int[]>();
int nTasks = 0;
// Read the data file. Each row specifies a driver cost, a count of tasks, and the task numbers
try
{
//parse parameters to extract data
//<line no> = bus driver no-1 (eg line 1 = driver0)
//1 4 3 4 16 17
//1 = cost of driver
// 4 = number of bus routes (tasks)
// 3, 4, 16, 17 = the bus routes (tasks) for the driver
var lines = Regex.Split(parameters, "\r\n|\r|\n");
foreach (var line in lines)
{
if (string.IsNullOrWhiteSpace(line))
{
continue;
}
int[] tasks;
driverCosts.Add(S.Constant(ReadDriver(line, out tasks)));
nTasks += tasks.Length;
Array.Sort <int>(tasks);
driversTasks.Add(tasks);
}
int nDrivers = driversTasks.Count;
// create a master list of unique tasks (bus routes) that be assigned to drivers
List <int> tasksU = new List <int>();
driversTasks.ForEach(x => tasksU.AddRange(x));
int[] allTasks = tasksU.OrderBy(x => x).Distinct().ToArray();
nTasks = allTasks.Length;
// We now have an array of all the tasks, and a list of all the drivers.
// The problem statement comes down to:
// - each task must be assigned exactly once
// - minimize the cost of drivers
// We add a boolean vector representing the drivers, true if the driver is to be used.
CspTerm[] driversUsed = S.CreateBooleanVector("drivers", nDrivers); // these are the Decision Variables
// We now create an array which maps which tasks are in which drivers.
// In addition to this static map, we create a dynamic map of the usage and the costs.
CspTerm[][] taskActualUse = new CspTerm[nTasks][];
CspTerm[] driverActualCost = new CspTerm[nDrivers];
for (int t = 0; t < nTasks; t++) //for each task / bus route
{
taskActualUse[t] = new CspTerm[nDrivers]; //for each route, array of all bus drivers (used to flag who may drive the route)
for (int r = 0; r < nDrivers; r++) //for each driver
{
taskActualUse[t][r] = (0 <= Array.BinarySearch <int>(driversTasks[r], allTasks[t])) ? driversUsed[r] : S.False;
}
S.AddConstraints(
S.ExactlyMofN(1, taskActualUse[t]) // this task appears exactly once
);
}
// set the goal: minimize total driver's cost
for (int r = 0; r < nDrivers; r++)
{
driverActualCost[r] = driversUsed[r] * driverCosts[r]; // dynamic cost map
}
S.TryAddMinimizationGoals(S.Sum(driverActualCost));
// now run the Solver and print the solutions
int solnId = 0;
ConstraintSolverSolution soln = S.Solve();
if (soln.HasFoundSolution)
{
System.Console.WriteLine("Solution #" + solnId++);
for (int d = 0; d < driversUsed.Length; d++)
{
object isUsed;
if (!soln.TryGetValue(driversUsed[d], out isUsed))
{
throw new InvalidProgramException("can't find drive in the solution: " + d.ToString());
}
// Take only the decision variables which turn out true.
// For each true row, print the row number and the list of tasks.
if (1 == (int)isUsed)
{
StringBuilder line = new StringBuilder(d.ToString());
line.Append(": ");
foreach (int x in driversTasks[d])
{
line.Append(x.ToString()).Append(", ");
}
System.Console.WriteLine(line.ToString());
}
}
}
if (solnId == 0)
{
System.Console.WriteLine("No solution found.");
}
}
catch (Exception e)
{
// Let the user know what went wrong.
Console.WriteLine("The file could not be read:");
Console.WriteLine(e.Message);
}
}
