public List <dtoAlgorithmAlternative> CalculateWeightedAverage(dtoAlgorithmInput input)
{
List <dtoAlgorithmAlternative> results = input.Alternatives;
double[] weights = input.Weights.ToArray();
//switch (input.NormalizeTo)
//{
// case NormalizeTo.simple:
// weights = weights.SimpleNormalize().ToArray();
// break;
// case NormalizeTo.standard:
weights = weights.Normalize().ToArray();
// break;
// case NormalizeTo.toOne:
// weights = weights.NormalizeTo1().ToArray();
// break;
//}
LM.MathLibrary.Algorithms.TriangularFuzzyWeightedAverage tp = new LM.MathLibrary.Algorithms.TriangularFuzzyWeightedAverage(input.Alternatives.Select(a => a.Values.ToArray()), weights);
TriangularFuzzyNumber[] values = tp.Elaborate();
if (values != null && values.Any())
{
long index = 0;
foreach (dtoAlgorithmAlternative alternative in results)
{
if (values[index] != null)
{
alternative.Ranking = values[index].Ranking();
alternative.FinalValue = values[index].ToCrispy();
alternative.FinalValueFuzzy = values[index].ToString();
alternative.IsFuzzyValue = true;
}
else
{
alternative.Ranking = 0;
alternative.FinalValue = 0;
alternative.FinalValueFuzzy = alternative.FinalValue.ToFuzzy().ToString();
alternative.IsFuzzyValue = true;
}
index++;
}
}
return(results);
}
public List <dtoAlgorithmAlternative> Calculate(dtoAlgorithmInput input)
{
List <dtoAlgorithmAlternative> results = input.Alternatives;
switch (input.Type)
{
case AlgorithmType.ahp:
break;
case AlgorithmType.owa:
results = CalculateOwa(input);
break;
case AlgorithmType.topsis:
results = CalculateTopsis(input);
break;
case AlgorithmType.weightedAverage:
results = CalculateWeightedAverage(input);
break;
}
return(results);
}
public List <dtoAlgorithmAlternative> CalculateTopsis(dtoAlgorithmInput input, IEnumerable <int> benefits = null, IEnumerable <int> costs = null)
{
List <dtoAlgorithmAlternative> results = input.Alternatives;
double[] weights = input.Weights.ToArray();
switch (input.NormalizeTo)
{
case NormalizeTo.simple:
weights = weights.SimpleNormalize().ToArray();
break;
case NormalizeTo.standard:
weights = weights.Normalize().ToArray();
break;
case NormalizeTo.toOne:
weights = weights.NormalizeTo1().ToArray();
break;
}
LM.MathLibrary.Algorithms.Topsis tp = new LM.MathLibrary.Algorithms.Topsis(input.Alternatives.Select(a => a.Values.ToArray()), weights, benefits, costs);
double[] values = tp.Elaborate();
if (values != null && values.Any())
{
values = (values.SumToOne()) ? values: values.NormalizeTo1().ToArray();
long index = 0;
foreach (dtoAlgorithmAlternative alternative in results)
{
alternative.Ranking = values[index];
alternative.FinalValue = values[index];
alternative.FinalValueFuzzy = new TriangularFuzzyNumber(alternative.FinalValue).ToString();
alternative.IsFuzzyValue = false;
index++;
}
}
return(results);
}
