discrete_optimization.generic_rcpsp_tools.solvers package

Subpackages

• discrete_optimization.generic_rcpsp_tools.solvers.lns_cp package
  • Submodules
  • discrete_optimization.generic_rcpsp_tools.solvers.lns_cp.neighbor_builder module
    • OptionNeighborRandom
      • OptionNeighborRandom.DEBUG
      • OptionNeighborRandom.LARGE
      • OptionNeighborRandom.MIX_ALL
      • OptionNeighborRandom.MIX_FAST
      • OptionNeighborRandom.MIX_LARGE_NEIGH
      • OptionNeighborRandom.NO_CONSTRAINT
    • Params
    • build_basic_random_and_neighbor()
    • build_basic_random_constraint_handler()
    • build_basic_time_window_constraint_handler()
    • build_constraint_handler_cut_part()
    • build_neighbor_mixing_cut_parts()
    • build_neighbor_mixing_methods()
    • build_neighbor_random()
    • constraint_neigh()
    • cut_parts()
    • mix()
    • mix_both()
    • mix_lot()
    • random_neigh()
    • return_random_basic_constraint_handler()
    • time_window_neigh()
  • discrete_optimization.generic_rcpsp_tools.solvers.lns_cp.neighbor_tools module
    • BasicConstraintBuilder
      • BasicConstraintBuilder.return_constraints()
    • ConstraintHandlerMultiskillAllocation
      • ConstraintHandlerMultiskillAllocation.adding_constraint_from_results_store()
    • ConstraintHandlerScheduling
      • ConstraintHandlerScheduling.adding_constraint_from_results_store()
    • EquilibrateMultiskillAllocation
      • EquilibrateMultiskillAllocation.adding_constraint_from_results_store()
    • EquilibrateMultiskillAllocationNonPreemptive
      • EquilibrateMultiskillAllocationNonPreemptive.adding_constraint_from_results_store()
    • NeighborBuilder
      • NeighborBuilder.find_subtasks()
    • NeighborBuilderMix
      • NeighborBuilderMix.find_subtasks()
    • NeighborBuilderSubPart
      • NeighborBuilderSubPart.find_subtasks()
    • NeighborBuilderTimeWindow
      • NeighborBuilderTimeWindow.find_subtasks()
    • NeighborConstraintBreaks
      • NeighborConstraintBreaks.find_subtasks()
    • NeighborRandom
      • NeighborRandom.find_subtasks()
    • NeighborRandomAndNeighborGraph
      • NeighborRandomAndNeighborGraph.find_subtasks()
    • ObjectiveSubproblem
      • ObjectiveSubproblem.GLOBAL_MAKESPAN
      • ObjectiveSubproblem.MAKESPAN_SUBTASKS
      • ObjectiveSubproblem.SUM_END_SUBTASKS
      • ObjectiveSubproblem.SUM_START_SUBTASKS
    • ParamsConstraintBuilder
      • ParamsConstraintBuilder.default()
      • ParamsConstraintBuilder.hyperparameters
    • constraint_number_of_change_in_worker_allocation()
    • constraint_number_of_change_in_worker_allocation_preemptive()
    • constraint_unit_used_subset_employees()
    • constraint_unit_used_subset_employees_preemptive()
    • constraint_unit_used_to_tasks()
    • constraint_unit_used_to_tasks_preemptive()
    • constraints_exchange_worker()
    • constraints_start_on_end()
    • constraints_start_on_end_preemptive()
    • constraints_strings()
    • constraints_strings_multiskill()
    • constraints_strings_multiskill_preemptive()
    • constraints_strings_preemptive()
    • get_max_time_solution()
    • intersect()
  • discrete_optimization.generic_rcpsp_tools.solvers.lns_cp.postpro_ls module
    • LocalSearchPostProcessSolution
      • LocalSearchPostProcessSolution.build_other_solution()
  • discrete_optimization.generic_rcpsp_tools.solvers.lns_cp.solution_repair module
    • NeighborRepairProblems
      • NeighborRepairProblems.adding_constraint_from_results_store()
    • compute_shift_extremities()
    • find_possible_problems_preemptive()
    • post_process_solution()
    • problem_constraints()
    • return_pauses_and_active_times()
  • discrete_optimization.generic_rcpsp_tools.solvers.lns_cp.solver module
    • ConstraintHandlerType
      • ConstraintHandlerType.MIX_SUBPROBLEMS
      • ConstraintHandlerType.SOLUTION_REPAIR
    • LnsCpMznGenericRcpspSolver
      • LnsCpMznGenericRcpspSolver.hyperparameters
    • build_constraint_handler()
    • build_constraint_handler_helper()
    • build_default_cp_model()
    • build_default_initial_solution()
    • build_default_postpro()
  • Module contents

Submodules

discrete_optimization.generic_rcpsp_tools.solvers.generic_rcpsp_solver module

class discrete_optimization.generic_rcpsp_tools.solvers.generic_rcpsp_solver.GenericRcpspSolver(problem: Problem, params_objective_function: ParamsObjectiveFunction | None = None, **kwargs: Any)

Bases: SolverDO

problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem

discrete_optimization.generic_rcpsp_tools.solvers.gphh module

class discrete_optimization.generic_rcpsp_tools.solvers.gphh.EvaluationGphh(value)

Bases: Enum

An enumeration.

PERMUTATION_DISTANCE = 1

SGS = 0

class discrete_optimization.generic_rcpsp_tools.solvers.gphh.FeatureEnum(value)

Bases: Enum

An enumeration.

ALL_DESCENDANTS = 'all_descendants'

DUMMY = 'DUMMY'

EARLIEST_FINISH_DATE = 'EFD'

EARLIEST_START_DATE = 'ESD'

LATEST_FINISH_DATE = 'LFD'

LATEST_START_DATE = 'LSD'

N_PREDECESSORS = 'n_predecessors'

N_SUCCESSORS = 'n_successors'

RESSOURCE_AVG = 'avg_res_requ'

RESSOURCE_MAX = 'max_res_requ'

RESSOURCE_MIN = 'min_res_requ'

RESSOURCE_NZ_MIN = 'nz_min_res_requ'

RESSOURCE_REQUIRED = 'res_requ'

RESSOURCE_TOTAL = 'total_nres'

TASK_DURATION = 'task_duration'

class discrete_optimization.generic_rcpsp_tools.solvers.gphh.GphhGenericRcpspSolver(problem: Problem, training_domains: list[Problem] | None = None, weight: int = 1, params_gphh: ParametersGphh | None = None, params_objective_function: ParamsObjectiveFunction | None = None, **kwargs)

Bases: GenericRcpspSolver

build_solution(domain, individual=None, func_heuristic=None)

evaluate_heuristic(individual, domains) → float

evaluation_method: EvaluationGphh

init_model()

Initialize internal model used to solve.

Can initialize a ortools, milp, gurobi, … model.

init_primitives(pset) → PrimitiveSet

initialize_cpm_data_for_training()

params_gphh: ParametersGphh

permutation_distance: PermutationDistance

pset: PrimitiveSet

reference_permutations: dict

solve(**kwargs)

Generic solving function.

Parameters:

• callbacks – list of callbacks used to hook into the various stage of the solve

• **kwargs – any argument specific to the solver

Solvers deriving from SolverDo should use callbacks methods .on_step_end(), … during solve(). But some solvers are not yet updated and are just ignoring it.

Returns (ResultStorage): a result object containing potentially a pool of solutions to a discrete-optimization problem

toolbox: Toolbox

training_domains: list[Problem]

weight: int

class discrete_optimization.generic_rcpsp_tools.solvers.gphh.ParametersGphh(set_feature, set_primitves, tournament_ratio, pop_size, n_gen, min_tree_depth, max_tree_depth, crossover_rate, mutation_rate, deap_verbose, evaluation, permutation_distance)

Bases: object

crossover_rate: float = None

deap_verbose: bool = None

static default()

static default_for_set_features(set_feature: set[FeatureEnum])

evaluation: EvaluationGphh = None

static fast_test()

max_tree_depth: int = None

min_tree_depth: int = None

static ms_default()

static ms_fast()

mutation_rate: float = None

n_gen: int = None

permutation_distance = 0

pop_size: int = None

set_feature: set[FeatureEnum] = None

set_primitves: PrimitiveSet = None

tournament_ratio: float = None

class discrete_optimization.generic_rcpsp_tools.solvers.gphh.PermutationDistance(value)

Bases: Enum

An enumeration.

HAMMING = 1

KTD = 0

KTD_HAMMING = 2

discrete_optimization.generic_rcpsp_tools.solvers.gphh.compute_cpm(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_all_descendants(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_average_resource_requirements(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_efd(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

Will only work if you store cpm results into the object. dirty trick

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_esd(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

Will only work if you store cpm results into the object. dirty trick

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_lfd(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

Will only work if you store cpm results into the object. dirty trick

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_lsd(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

Will only work if you store cpm results into the object. dirty trick

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_maximum_resource_requirements(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_minimum_resource_requirements(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_n_predecessors(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_n_successors(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_non_zero_minimum_resource_requirements(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_resource_requirements(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_task_duration(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.feature_total_n_res(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.get_dummy(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, cpm, cpm_esd, task_id: int, increase: int, **kwargs)

Will only work if you store cpm results into the object. dirty trick

discrete_optimization.generic_rcpsp_tools.solvers.gphh.get_resource_requirements_across_duration(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, task_id: int, **kwargs)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.if_then_else(input1, output1, output2)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.max_operator(left, right)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.min_operator(left, right)

discrete_optimization.generic_rcpsp_tools.solvers.gphh.protected_div(left, right)

discrete_optimization.generic_rcpsp_tools.solvers.ls module

class discrete_optimization.generic_rcpsp_tools.solvers.ls.LsGenericRcpspSolver(problem: RcpspProblem | PreemptiveRcpspProblem | SpecialConstraintsPreemptiveRcpspProblem | MultiskillRcpspProblem | VariantMultiskillRcpspProblem, params_objective_function: ParamsObjectiveFunction | None = None, ls_solver: LsSolverType = LsSolverType.SA, **args)

Bases: GenericRcpspSolver

hyperparameters: list[Hyperparameter] = [CategoricalHyperparameter(name='init_solution_process', default=False, depends_on=None, name_in_kwargs='init_solution_process'), EnumHyperparameter(name='ls_solver', default=<LsSolverType.SA: 0>, depends_on=None, name_in_kwargs='ls_solver'), FloatHyperparameter(name='temperature', default=3, depends_on=('ls_solver', [<LsSolverType.SA: 0>]), name_in_kwargs='temperature', low=0.01, high=10, suggest_low=False, suggest_high=False, step=None, log=False), IntegerHyperparameter(name='nb_iteration_no_improvement', default=200, depends_on=None, name_in_kwargs='nb_iteration_no_improvement', low=10, high=2000, step=1, log=False)]

Hyperparameters available for this solver.

These hyperparameters are to be feed to **kwargs found in

• __init__()

• init_model() (when available)

• solve()

solve(callbacks: list[Callback] | None = None, **kwargs)

Generic solving function.

Parameters:

• callbacks – list of callbacks used to hook into the various stage of the solve

• **kwargs – any argument specific to the solver

Solvers deriving from SolverDo should use callbacks methods .on_step_end(), … during solve(). But some solvers are not yet updated and are just ignoring it.

Returns (ResultStorage): a result object containing potentially a pool of solutions to a discrete-optimization problem

class discrete_optimization.generic_rcpsp_tools.solvers.ls.LsSolverType(value)

Bases: Enum

An enumeration.

HC = 1

SA = 0

Module contents