Source code for discrete_optimization.fjsp.problem

#  Copyright (c) 2024 AIRBUS and its affiliates.
#  This source code is licensed under the MIT license found in the
#  LICENSE file in the root directory of this source tree.

from __future__ import annotations

import logging
from dataclasses import dataclass

from discrete_optimization.generic_tasks_tools.multimode import (
    MultimodeProblem,
    MultimodeSolution,
)
from discrete_optimization.generic_tasks_tools.precedence import PrecedenceProblem
from discrete_optimization.generic_tasks_tools.scheduling import (
    SchedulingProblem,
    SchedulingSolution,
)
from discrete_optimization.generic_tools.do_problem import (
    EncodingRegister,
    ModeOptim,
    ObjectiveDoc,
    ObjectiveHandling,
    ObjectiveRegister,
    Solution,
    TypeObjective,
)
from discrete_optimization.jsp.problem import Subjob, Task

logger = logging.getLogger(__name__)


[docs] class FJobShopSolution(SchedulingSolution[Task], MultimodeSolution[Task]): problem: FJobShopProblem def __init__( self, problem: FJobShopProblem, schedule: list[list[tuple[int, int, int, int]]] ): # For each job and sub-job, start, end time, machine id, and option choice given as tuple of int. self.problem = problem self.schedule = schedule
[docs] def copy(self) -> FJobShopSolution: return FJobShopSolution(problem=self.problem, schedule=self.schedule)
[docs] def change_problem(self, new_problem: FJobShopProblem) -> None: self.problem = new_problem
[docs] def get_end_time(self, task: Task) -> int: j, k = task return self.schedule[j][k][1]
[docs] def get_start_time(self, task: Task) -> int: j, k = task return self.schedule[j][k][0]
[docs] def get_machine(self, task: Task) -> int: j, k = task return self.schedule[j][k][2]
[docs] def get_mode(self, task: Task) -> int: """Get 'mode' of given task, aka chosen machine.""" j, k = task return self.schedule[j][k][-1]
SubjobOptions = list[Subjob]
[docs] @dataclass class Job: job_id: int sub_jobs: list[SubjobOptions]
[docs] class FJobShopProblem( SchedulingProblem[Task], MultimodeProblem[Task], PrecedenceProblem[Task] ): n_jobs: int n_machines: int list_jobs: list[Job] def __init__( self, list_jobs: list[Job], n_jobs: int = None, n_machines: int = None, horizon: int = None, ): self.list_jobs = list_jobs self.n_jobs = n_jobs self.n_machines = n_machines self.list_jobs = list_jobs if self.n_jobs is None: self.n_jobs = len(list_jobs) if self.n_machines is None: self.n_machines = len( set( [ option.machine_id for job in self.list_jobs for options in job.sub_jobs for option in options ] ) ) self.n_all_jobs = sum(len(subjob.sub_jobs) for subjob in self.list_jobs) self.job_per_machines = {i: [] for i in range(self.n_machines)} for k in range(self.n_jobs): for sub_k in range(len(list_jobs[k].sub_jobs)): for option in range(len(list_jobs[k].sub_jobs[sub_k])): self.job_per_machines[ list_jobs[k].sub_jobs[sub_k][option].machine_id ] += [(k, sub_k, option)] self.horizon = horizon if self.horizon is None: self.horizon = sum( sum( max(subjob.processing_time for subjob in subjob_opt) for subjob_opt in job.sub_jobs ) for job in self.list_jobs ) self.nb_subjob_per_job = { i: len(self.list_jobs[i].sub_jobs) for i in range(self.n_jobs) } self.subjob_possible_machines = { (i, j): set(x.machine_id for x in self.list_jobs[i].sub_jobs[j]) for i in range(self.n_jobs) for j in range(self.nb_subjob_per_job[i]) } self.duration_per_machines = { (i, j): { x.machine_id: x.processing_time for x in self.list_jobs[i].sub_jobs[j] } for (i, j) in self.subjob_possible_machines }
[docs] def get_makespan_upper_bound(self) -> int: return self.horizon
@property def tasks_list(self) -> list[Task]: return [ (j, k) for j, job in enumerate(self.list_jobs) for k in range(len(job.sub_jobs)) ]
[docs] def get_precedence_constraints(self) -> dict[Task, list[Task]]: return { (j, k): [(j, k + 1)] if k + 1 < len(job.sub_jobs) else [] for j, job in enumerate(self.list_jobs) for k in range(len(job.sub_jobs)) }
[docs] def get_task_modes(self, task: Task) -> set[int]: j, k = task return set(range(len(self.list_jobs[j].sub_jobs[k])))
[docs] def get_last_tasks(self) -> list[Task]: return [(j, len(job.sub_jobs) - 1) for j, job in enumerate(self.list_jobs)]
[docs] def evaluate(self, variable: FJobShopSolution) -> dict[str, float]: return {"makespan": variable.get_max_end_time()}
[docs] def satisfy(self, variable: FJobShopSolution) -> bool: if not all( variable.get_machine(task=task) in machines for task, machines in self.subjob_possible_machines.items() ): logger.info("Unallowed machine used for some subjob") return False for m in self.job_per_machines: sorted_ = sorted( [ variable.schedule[x[0]][x[1]] for x in self.job_per_machines[m] if variable.schedule[x[0]][x[1]][2] == m ], key=lambda y: y[0], ) len_ = len(sorted_) for i in range(1, len_): if sorted_[i][0] < sorted_[i - 1][1]: logger.info("Overlapping task on same machines") return False for job in range(self.n_jobs): s_j = 0 i_opt = variable.schedule[job][s_j][-1] machine_id = variable.schedule[job][s_j][2] if self.list_jobs[job].sub_jobs[s_j][i_opt].machine_id != machine_id: logger.info( f"Machine choice and option choice does not match for task {job, s_j}." ) return False if not ( variable.schedule[job][s_j][1] - variable.schedule[job][s_j][0] == self.duration_per_machines[(job, s_j)][machine_id] ): logger.info( f"Duration of task {job, s_j} not coherent with the machine choice " ) for s_j in range(1, len(variable.schedule[job])): if variable.schedule[job][s_j][0] < variable.schedule[job][s_j - 1][1]: logger.info( f"Precedence constraint not respected between {job, s_j}" f"and {job, s_j - 1}" ) return False machine_id = variable.schedule[job][s_j][2] if not ( variable.schedule[job][s_j][1] - variable.schedule[job][s_j][0] == self.duration_per_machines[(job, s_j)][machine_id] ): logger.info( f"Duration of task {job, s_j} not coherent with the machine choice " ) return False i_opt = variable.schedule[job][s_j][-1] if self.list_jobs[job].sub_jobs[s_j][i_opt].machine_id != machine_id: logger.info( f"Machine choice and option choice does not match for task {job, s_j}." ) return False return True
[docs] def get_attribute_register(self) -> EncodingRegister: return EncodingRegister(dict_attribute_to_type={})
[docs] def get_solution_type(self) -> type[Solution]: return FJobShopSolution
[docs] def get_objective_register(self) -> ObjectiveRegister: return ObjectiveRegister( dict_objective_to_doc={ "makespan": ObjectiveDoc(type=TypeObjective.OBJECTIVE, default_weight=1) }, objective_sense=ModeOptim.MINIMIZATION, objective_handling=ObjectiveHandling.AGGREGATE, )