Supervised Learning of the Optimal Objective Function Value in Chemical Production Scheduling

Mixed-integer programming (MIP) can be used to formulate and solve complex production scheduling problems in the field of process systems engineering. However, the solution of MIP models may require a long computing time due to the combinatorial complexity of the problems. In this work, we propose s...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Teemu J. Ikonen (20724395) (author)
مؤلفون آخرون: Boeun Kim (1234278) (author), Christos T. Maravelias (206037) (author), Iiro Harjunkoski (2331031) (author)
منشور في: 2025
الموضوعات:
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1852022711690199040
author Teemu J. Ikonen (20724395)
author2 Boeun Kim (1234278)
Christos T. Maravelias (206037)
Iiro Harjunkoski (2331031)
author2_role author
author
author
author_facet Teemu J. Ikonen (20724395)
Boeun Kim (1234278)
Christos T. Maravelias (206037)
Iiro Harjunkoski (2331031)
author_role author
dc.creator.none.fl_str_mv Teemu J. Ikonen (20724395)
Boeun Kim (1234278)
Christos T. Maravelias (206037)
Iiro Harjunkoski (2331031)
dc.date.none.fl_str_mv 2025-02-14T04:45:57Z
dc.identifier.none.fl_str_mv 10.1021/acs.iecr.4c03045.s002
dc.relation.none.fl_str_mv https://figshare.com/articles/dataset/Supervised_Learning_of_the_Optimal_Objective_Function_Value_in_Chemical_Production_Scheduling/28415843
dc.rights.none.fl_str_mv CC BY-NC 4.0
info:eu-repo/semantics/openAccess
dc.subject.none.fl_str_mv Biochemistry
Biological Sciences not elsewhere classified
Mathematical Sciences not elsewhere classified
Information Systems not elsewhere classified
studied objective functions
process systems engineering
predictions allow us
different problem features
2 </ sup
best prediction models
prediction models
r </
>< sup
supervised learning
makespan minimization
integer programming
instance parameters
high 
four classes
combinatorial complexity
dc.title.none.fl_str_mv Supervised Learning of the Optimal Objective Function Value in Chemical Production Scheduling
dc.type.none.fl_str_mv Dataset
info:eu-repo/semantics/publishedVersion
dataset
description Mixed-integer programming (MIP) can be used to formulate and solve complex production scheduling problems in the field of process systems engineering. However, the solution of MIP models may require a long computing time due to the combinatorial complexity of the problems. In this work, we propose supervised learning models to predict the optimal objective function value on four classes of scheduling problems, which can be useful in a number of settings. To improve the accuracy of the prediction models, we device a number of machine learning features based on the instance parameters. The studied objective functions are cost and makespan minimization. Based on the results, the prediction accuracy is highthe coefficients of determination with the best prediction models are <i>r</i><sup>2</sup> > 0.97 on the four classes of problems. These predictions allow us to predict how different problem features (e.g., new orders or disturbances) affect the optimal objective function value.
eu_rights_str_mv openAccess
id Manara_1ebcbdb21bbd5767ea8e8cd323bcdaad
identifier_str_mv 10.1021/acs.iecr.4c03045.s002
network_acronym_str Manara
network_name_str ManaraRepo
oai_identifier_str oai:figshare.com:article/28415843
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
rights_invalid_str_mv CC BY-NC 4.0
spelling Supervised Learning of the Optimal Objective Function Value in Chemical Production SchedulingTeemu J. Ikonen (20724395)Boeun Kim (1234278)Christos T. Maravelias (206037)Iiro Harjunkoski (2331031)BiochemistryBiological Sciences not elsewhere classifiedMathematical Sciences not elsewhere classifiedInformation Systems not elsewhere classifiedstudied objective functionsprocess systems engineeringpredictions allow usdifferent problem features2 </ supbest prediction modelsprediction modelsr </>< supsupervised learningmakespan minimizationinteger programminginstance parametershigh four classescombinatorial complexityMixed-integer programming (MIP) can be used to formulate and solve complex production scheduling problems in the field of process systems engineering. However, the solution of MIP models may require a long computing time due to the combinatorial complexity of the problems. In this work, we propose supervised learning models to predict the optimal objective function value on four classes of scheduling problems, which can be useful in a number of settings. To improve the accuracy of the prediction models, we device a number of machine learning features based on the instance parameters. The studied objective functions are cost and makespan minimization. Based on the results, the prediction accuracy is highthe coefficients of determination with the best prediction models are <i>r</i><sup>2</sup> > 0.97 on the four classes of problems. These predictions allow us to predict how different problem features (e.g., new orders or disturbances) affect the optimal objective function value.2025-02-14T04:45:57ZDatasetinfo:eu-repo/semantics/publishedVersiondataset10.1021/acs.iecr.4c03045.s002https://figshare.com/articles/dataset/Supervised_Learning_of_the_Optimal_Objective_Function_Value_in_Chemical_Production_Scheduling/28415843CC BY-NC 4.0info:eu-repo/semantics/openAccessoai:figshare.com:article/284158432025-02-14T04:45:57Z
spellingShingle Supervised Learning of the Optimal Objective Function Value in Chemical Production Scheduling
Teemu J. Ikonen (20724395)
Biochemistry
Biological Sciences not elsewhere classified
Mathematical Sciences not elsewhere classified
Information Systems not elsewhere classified
studied objective functions
process systems engineering
predictions allow us
different problem features
2 </ sup
best prediction models
prediction models
r </
>< sup
supervised learning
makespan minimization
integer programming
instance parameters
high 
four classes
combinatorial complexity
status_str publishedVersion
title Supervised Learning of the Optimal Objective Function Value in Chemical Production Scheduling
title_full Supervised Learning of the Optimal Objective Function Value in Chemical Production Scheduling
title_fullStr Supervised Learning of the Optimal Objective Function Value in Chemical Production Scheduling
title_full_unstemmed Supervised Learning of the Optimal Objective Function Value in Chemical Production Scheduling
title_short Supervised Learning of the Optimal Objective Function Value in Chemical Production Scheduling
title_sort Supervised Learning of the Optimal Objective Function Value in Chemical Production Scheduling
topic Biochemistry
Biological Sciences not elsewhere classified
Mathematical Sciences not elsewhere classified
Information Systems not elsewhere classified
studied objective functions
process systems engineering
predictions allow us
different problem features
2 </ sup
best prediction models
prediction models
r </
>< sup
supervised learning
makespan minimization
integer programming
instance parameters
high 
four classes
combinatorial complexity