mcfadd / job_shop_schedule_problem Goto Github PK

Parallel Tabu Search and Genetic Algorithm for the Job Shop Schedule Problem with Sequence Dependent Set Up Times

License: ISC License

Python 97.81% HTML 2.19%

job-scheduler job-shop-scheduling-problem job-shop-schedulling tabu-search genetic-algorithm np-hard combinatorics combinatorial-optimization python cython

job_shop_schedule_problem's People

Contributors

Stargazers

Watchers

job_shop_schedule_problem's Issues

TODO List

This issue is for recording a general TODO list of tasks.

Todo

Flow diagram describing the steps of our program.
Create solution class in data_set.py
Create feasible_solution_factory.py for generating feasible solutions.
Finish tabu_search.
Create another instance of the problem (i.e. jobTasks.csv, sequenceDependencyMatrix.csv, machineRunSpeed.csv) that is slightly bigger than the one currently in the data
Create a function that generates schedules with start/end times for the job-tasks on each machines given an operation list. (see code TODO comment in Solution class)
Research how to run multiple TS processes in parallel (i.e. multi-programming) and collect their results when they finish. (see multi-programming in python)
Research how/if we can use Cython to speed up runtime (i.e. # iterations of TS performed in a certain duration)

Ask for help

I want to use it to solve the aircraft scheduling problem problems on the basis of your program, so I must have a good understanding of your case. Can you recommend some articles?

Understand the input data

Could you explain the "pieces" parameter in class "Task"? Is this like the product quantities of a task? What it is used for in the algorithm?

JSSP.solution._makespan

there are some problem with packets versions
i upgraded them and changed setup.py requirements but when i start the example script is says that JSSP.solution._makespan cant be found

Benchmarks

This issue is for recording benchmark stuff such as average number of iterations tabu search performs in a certain duration. The benchmarks are obtained from benchmark.py.

How to understanding sequenceDependencyMatrix.csv matrix?

Thank you for your job, it is amazing！
Could you explain the ”seq dep matrix“? What it is used for in the algorithm?

Need some intution about sequence dependency matrix

Error running parallel tabu search on Windows

After building and installing JSSP-0.1.0.win-amd64.zip (Windows build), then running TS_Example.ipynb, the following error appeared:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-2-8458fe138112> in <module>
     34                                    benchmark=benchmark,
     35                                    verbose=verbose,
---> 36                                    progress_bar=progress_bar
     37                                   )
     38 

~\Anaconda3\envs\JSSP_env\lib\site-packages\JSSP\solver.py in tabu_search_time(self, runtime, num_solutions_per_process, num_processes, tabu_list_size, neighborhood_size, neighborhood_wait, probability_change_machine, reset_threshold, initial_solutions, benchmark, verbose, progress_bar)
     70                                  probability_change_machine=probability_change_machine,
     71                                  reset_threshold=reset_threshold, initial_solutions=initial_solutions,
---> 72                                  benchmark=benchmark, verbose=verbose, progress_bar=progress_bar)
     73 
     74     def tabu_search_iter(self, iterations, num_solutions_per_process=1, num_processes=4, tabu_list_size=50,

~\Anaconda3\envs\JSSP_env\lib\site-packages\JSSP\solver.py in _tabu_search(self, stopping_condition, time_condition, num_solutions_per_process, num_processes, tabu_list_size, neighborhood_size, neighborhood_wait, probability_change_machine, reset_threshold, initial_solutions, benchmark, verbose, progress_bar)
    186         # start child processes
    187         for p in processes:
--> 188             p.start()
    189             if verbose:
    190                 print(f"child TS process started. pid = {p.pid}")

~\Anaconda3\envs\JSSP_env\lib\multiprocessing\process.py in start(self)
    103                'daemonic processes are not allowed to have children'
    104         _cleanup()
--> 105         self._popen = self._Popen(self)
    106         self._sentinel = self._popen.sentinel
    107         # Avoid a refcycle if the target function holds an indirect

~\Anaconda3\envs\JSSP_env\lib\multiprocessing\context.py in _Popen(process_obj)
    221     @staticmethod
    222     def _Popen(process_obj):
--> 223         return _default_context.get_context().Process._Popen(process_obj)
    224 
    225 class DefaultContext(BaseContext):

~\Anaconda3\envs\JSSP_env\lib\multiprocessing\context.py in _Popen(process_obj)
    320         def _Popen(process_obj):
    321             from .popen_spawn_win32 import Popen
--> 322             return Popen(process_obj)
    323 
    324     class SpawnContext(BaseContext):

~\Anaconda3\envs\JSSP_env\lib\multiprocessing\popen_spawn_win32.py in __init__(self, process_obj)
     63             try:
     64                 reduction.dump(prep_data, to_child)
---> 65                 reduction.dump(process_obj, to_child)
     66             finally:
     67                 set_spawning_popen(None)

~\Anaconda3\envs\JSSP_env\lib\multiprocessing\reduction.py in dump(obj, file, protocol)
     58 def dump(obj, file, protocol=None):
     59     '''Replacement for pickle.dump() using ForkingPickler.'''
---> 60     ForkingPickler(file, protocol).dump(obj)
     61 
     62 #

~\Anaconda3\envs\JSSP_env\lib\site-packages\JSSP\solution\makespan.cp36-win_amd64.pyd in View.MemoryView._memoryviewslice.__reduce_cython__()

TypeError: no default __reduce__ due to non-trivial __cinit__

The problem has to do with pickling cython memory views. When python.multiprocessing.Process starts (i.e. is spawned), the memory views in makespan.cp36-win_amd64.pyd cannot be pickled.

Possible solutions

create a __reduce__ function that converts the memory views to a numpy arrays as described in this stackoverflow thread

Helpful Resources

This issue contains links to resources for using git version control, writing documentation in markdown, and writing code in python among other things.

Git

Markdown

Python

Diagrams

Cython

errors

It gives many errors during the installation phase of the package. Python-dev is also not working.

mcfadd / job_shop_schedule_problem Goto Github PK

job_shop_schedule_problem's People

Contributors

Stargazers

Watchers

Forkers

job_shop_schedule_problem's Issues

Todo

Possible solutions

Recommend Projects

Recommend Topics

Recommend Org