/// <summary>
/// Searches for the city that can be placed best with the least increase in cost.
/// </summary>
/// <param name="problem"></param>
/// <param name="calculator"></param>
/// <param name="genomes"></param>
/// <param name="cities_to_place"></param>
/// <returns></returns>
public BestPlacementResult CalculateBestPlacement(
GeneticProblem problem,
FitnessCalculator calculator,
List <int> genomes,
List <int> cities_to_place)
{
// initialize the best placement result.
BestPlacementResult result
= new BestPlacementResult();
// try and place all cities.
for (int city_idx = 0; city_idx < cities_to_place.Count; city_idx++)
{
// try to place first city.
int city = cities_to_place[city_idx];
// place the city and check the result.
BestPlacementResult current_result =
this.CalculateBestPlacement(
problem,
calculator,
genomes,
city);
// check current result
if (current_result.Increase < result.Increase)
{
result = current_result;
}
}
return(result);
}
/// <summary>
/// Calculates and places this city in the least cost position.
/// </summary>
/// <param name="problem"></param>
/// <param name="calculator"></param>
/// <param name="genomes"></param>
/// <param name="city_to_place"></param>
public void Do(GeneticProblem problem, FitnessCalculator calculator, List <int> genomes, int city_to_place)
{
// calculate the result.
BestPlacementResult result = this.CalculateBestPlacement(problem, calculator, genomes, city_to_place);
// do the placement.
this.Do(genomes, result);
}
/// <summary>
/// Calculates and places the given cities one by one in the least cost position.
/// </summary>
/// <param name="problem"></param>
/// <param name="calculator"></param>
/// <param name="genomes"></param>
/// <param name="cities_to_place"></param>
public void Do(GeneticProblem problem, FitnessCalculator calculator,
List <int> genomes, List <int> cities_to_place)
{
this.RaiseNewStart();
while (cities_to_place.Count > 0)
{
// calculate the result.
BestPlacementResult result = this.CalculateBestPlacement(problem, calculator, genomes, cities_to_place);
// remove from the cities to place list.
cities_to_place.Remove(result.City);
// do the placement.
this.Do(genomes, result);
}
}
/// <summary>
/// Calculates and places the given cities by randomly selecting them and placing them into the least cost position.
/// </summary>
/// <param name="problem"></param>
/// <param name="calculator"></param>
/// <param name="genomes"></param>
/// <param name="cities_to_place"></param>
public void DoFast(GeneticProblem problem, FitnessCalculator calculator, List <int> genomes, List <int> cities_to_place)
{
this.RaiseNewStart();
while (cities_to_place.Count > 0)
{
// select a random city.
int random_idx = StaticRandomGenerator.Get().Generate(cities_to_place.Count);
int city = cities_to_place[random_idx];
this.RaiseNewCity(city);
// remove the selected city.
cities_to_place.RemoveAt(random_idx);
// calculate the result.
BestPlacementResult result = this.CalculateBestPlacement(problem, calculator, genomes, city);
// do the placement.
this.Do(genomes, result);
this.RaiseNewRoute(genomes);
}
}
/// <summary>
/// Searches for the best place to insert the given city.
/// </summary>
/// <param name="problem"></param>
/// <param name="calculator"></param>
/// <param name="genomes"></param>
/// <param name="city_to_place"></param>
/// <returns></returns>
public BestPlacementResult CalculateBestPlacement(
GeneticProblem problem,
FitnessCalculator calculator,
List <int> genomes,
int city_to_place)
{ // initialize the best placement result.
BestPlacementResult result
= new BestPlacementResult();
result.City = city_to_place;
// initialize the best increase.
double increase = 0;
if (genomes.Count > 0)
{
// calculate from first.
double new_weights =
problem.Weight(problem.First, city_to_place)
+ (problem.Weight(city_to_place, genomes[0]));
// calculate the old weights.
double old_weight =
problem.Weight(problem.First, genomes[0]);
// calculate the difference to know the increase.
double new_increase =
new_weights - (old_weight);
// set the new increase.
increase = new_increase;
// set the result city.
result.CityIdx = 0;
for (int idx = 0; idx < genomes.Count - 1; idx++)
{
// calculate the new weights.
new_weights =
problem.Weight(genomes[idx], city_to_place)
+ (problem.Weight(city_to_place, genomes[idx + 1]));
// calculate the old weights.
old_weight =
problem.Weight(genomes[idx], genomes[idx + 1]);
// calculate the difference to know the increase.
new_increase =
new_weights - (old_weight);
//if (increase == null || new_increase < increase)
if (new_increase < increase)
{
// set the new increase.
increase = new_increase;
// set the result city.
result.CityIdx = idx + 1;
}
}
// test to the last.
// calculate the new weights.
new_weights =
problem.Weight(genomes[genomes.Count - 1], city_to_place)
+ (problem.Weight(city_to_place, problem.Last));
// calculate the old weights.
old_weight =
problem.Weight(genomes[genomes.Count - 1], problem.Last);
// calculate the difference to know the increase.
new_increase =
new_weights - (old_weight);
//if (increase == null || new_increase < increase)
if (new_increase < increase)
{
// set the new increase.
increase = new_increase;
// set the result city.
result.CityIdx = genomes.Count;
}
}
else
{
// calculate the new weights.
double new_weights =
problem.Weight(problem.First, city_to_place)
+ (problem.Weight(city_to_place, problem.Last));
// calculate the old weights.
double old_weight =
problem.Weight(problem.First, problem.Last);
// calculate the difference to know the increase.
double new_increase =
new_weights - (old_weight);
// set the new increase.
increase = new_increase;
// set the result city.
result.CityIdx = 0;
}
// calculate the fitness.
result.Increase = increase;
// return result.
return(result);
}
/// <summary>
/// Searches for the best place to insert the given city.
/// </summary>
/// <param name="problem"></param>
/// <param name="calculator"></param>
/// <param name="genomes"></param>
/// <param name="city_to_place"></param>
/// <returns></returns>
public BestPlacementResult CalculateBestPlacement(
GeneticProblem problem,
FitnessCalculator calculator,
List<int> genomes,
int city_to_place)
{
// initialize the best placement result.
BestPlacementResult result
= new BestPlacementResult();
result.City = city_to_place;
// initialize the best increase.
double increase = 0;
if (genomes.Count > 0)
{
// calculate from first.
double new_weights =
problem.Weight(problem.First, city_to_place)
+ (problem.Weight(city_to_place, genomes[0]));
// calculate the old weights.
double old_weight =
problem.Weight(problem.First, genomes[0]);
// calculate the difference to know the increase.
double new_increase =
new_weights - (old_weight);
// set the new increase.
increase = new_increase;
// set the result city.
result.CityIdx = 0;
for (int idx = 0; idx < genomes.Count - 1; idx++)
{
// calculate the new weights.
new_weights =
problem.Weight(genomes[idx], city_to_place)
+ (problem.Weight(city_to_place, genomes[idx + 1]));
// calculate the old weights.
old_weight =
problem.Weight(genomes[idx], genomes[idx + 1]);
// calculate the difference to know the increase.
new_increase =
new_weights - (old_weight);
//if (increase == null || new_increase < increase)
if (new_increase < increase)
{
// set the new increase.
increase = new_increase;
// set the result city.
result.CityIdx = idx + 1;
}
}
// test to the last.
// calculate the new weights.
new_weights =
problem.Weight(genomes[genomes.Count - 1], city_to_place)
+ (problem.Weight(city_to_place, problem.Last));
// calculate the old weights.
old_weight =
problem.Weight(genomes[genomes.Count - 1], problem.Last);
// calculate the difference to know the increase.
new_increase =
new_weights - (old_weight);
//if (increase == null || new_increase < increase)
if (new_increase < increase)
{
// set the new increase.
increase = new_increase;
// set the result city.
result.CityIdx = genomes.Count;
}
}
else
{
// calculate the new weights.
double new_weights =
problem.Weight(problem.First, city_to_place)
+ (problem.Weight(city_to_place, problem.Last));
// calculate the old weights.
double old_weight =
problem.Weight(problem.First, problem.Last);
// calculate the difference to know the increase.
double new_increase =
new_weights - (old_weight);
// set the new increase.
increase = new_increase;
// set the result city.
result.CityIdx = 0;
}
// calculate the fitness.
result.Increase = increase;
// return result.
return result;
}
/// <summary>
/// Calculates and places the given cities by randomly selecting them and placing them into the least cost position.
/// </summary>
/// <param name="problem"></param>
/// <param name="calculator"></param>
/// <param name="genomes"></param>
/// <param name="cities_to_place"></param>
public void DoFast(GeneticProblem problem, FitnessCalculator calculator, List<int> genomes, List<int> cities_to_place)
{
this.RaiseNewStart();
while (cities_to_place.Count > 0)
{
// select a random city.
int random_idx = StaticRandomGenerator.Get().Generate(cities_to_place.Count);
int city = cities_to_place[random_idx];
this.RaiseNewCity(city);
// remove the selected city.
cities_to_place.RemoveAt(random_idx);
// calculate the result.
BestPlacementResult result = this.CalculateBestPlacement(problem, calculator, genomes, city);
// do the placement.
this.Do(genomes, result);
this.RaiseNewRoute(genomes);
}
}
/// <summary>
/// Calculates and places the given cities one by one in the least cost position.
/// </summary>
/// <param name="problem"></param>
/// <param name="calculator"></param>
/// <param name="genomes"></param>
/// <param name="cities_to_place"></param>
public void Do(GeneticProblem problem, FitnessCalculator calculator,
List<int> genomes, List<int> cities_to_place)
{
this.RaiseNewStart();
while (cities_to_place.Count > 0)
{
// calculate the result.
BestPlacementResult result = this.CalculateBestPlacement(problem, calculator, genomes, cities_to_place);
// remove from the cities to place list.
cities_to_place.Remove(result.City);
// do the placement.
this.Do(genomes, result);
}
}
/// <summary>
/// Calculates and places this city in the least cost position.
/// </summary>
/// <param name="problem"></param>
/// <param name="calculator"></param>
/// <param name="genomes"></param>
/// <param name="city_to_place"></param>
public void Do(GeneticProblem problem, FitnessCalculator calculator, List<int> genomes, int city_to_place)
{
// calculate the result.
BestPlacementResult result = this.CalculateBestPlacement(problem, calculator, genomes, city_to_place);
// do the placement.
this.Do(genomes, result);
}
/// <summary>
/// Searches for the city that can be placed best with the least increase in cost.
/// </summary>
/// <param name="problem"></param>
/// <param name="calculator"></param>
/// <param name="genomes"></param>
/// <param name="cities_to_place"></param>
/// <returns></returns>
public BestPlacementResult CalculateBestPlacement(
GeneticProblem problem,
FitnessCalculator calculator,
List<int> genomes,
List<int> cities_to_place)
{
// initialize the best placement result.
BestPlacementResult result
= new BestPlacementResult();
// try and place all cities.
for (int city_idx = 0; city_idx < cities_to_place.Count; city_idx++)
{
// try to place first city.
int city = cities_to_place[city_idx];
// place the city and check the result.
BestPlacementResult current_result =
this.CalculateBestPlacement(
problem,
calculator,
genomes,
city);
// check current result
if (current_result.Increase < result.Increase)
{
result = current_result;
}
}
return result;
}
