Seite - 83 - in Algorithms for Scheduling Problems
Bild der Seite - 83 -
Text der Seite - 83 -
Algorithms 2018,11, 68
of tasks;eachset isassignedaprocessingorder (Proc.Order). Thesesetsare formedaccordingto the
processingcompatibility.
݉ͳ ݉ʹ
Ordered 1 2 1 2 Proc. Order
2 3 13 15 5 6 7 8 1 4 9 10 11 12 14 16 Batch
1 4 2 1 3 4 Jobs
Figure7.Chromosomerepresentationexamplewith twomachinesperstage.
݉ͳ ݉ʹ ݉͵ ݉Ͷ
Ordered 1 2 1 1 Proc. Order
2 3 6 7 3 15 5 8 10 11 12 16 1 4 9 14 Batch
1 4 2 1 4 2 3 4 1 3 4 Jobs
Figure8.Chromosomerepresentationexamplewith fourmachinesperstage.
The tasksof Jobs1,3,and4canbeprocessed in thesamesetat thesametime. Thetasksof Job2
areput intoaseparatesetbecause theyarenotcompatiblewitheachother. Forexample, inFigure7,
the first two tasks (1-1, 4-1) are processed byMachine 1, then the second setwith one task (2-1).
InMachine2, twotasksof thefirst set (1-2and3-1)areprocessed,assumingthat themachinehasonly
thecapacity toprocessupto6batches (60kg). Thesecondsetwill thenbefollowedbyonetask (4-1).
Figure8showsfourmachinesperstage. Fourbatchesareassignedtoeachmachine.Machines1
and2process twosetsof tasks.Machines3and4processonlyoneset.Uponcompletionof the tasks
processed insets inStep1, in the laterstages, theywillbeprocessed individually.
Ordercrossover (OX) [32,34–37]: Twocutpointsareselectedrandomly; thepartof thefirstparent
locatedbetween these twocutpoints is copied to the child. Thechild’sother spacesare left blank.
Thevaluescopiedto thechildaredeleted inthesecondparent. Thefollowingpositions in thechild
arefilledstarting in theblankspacesandconsideringtheorder foundin thesecondparent (Figure9).
Figure9.Ordercrossover (OX)operator.
5.3.MutationOperators
Mutationoperatorsproducesmallchanges inindividualsaccordingtoaprobability. Thisoperator
helps to prevent falling into local optima and to extend the search space of the algorithm. Three
mutationoperatorsareconsidered:Displacement,Exchange,andInsertion.
Displacement isageneralizationof the insertionmutation, inwhich insteadofmovingasingle
value, severalvaluesarechanged(Figure10). Theexchangeoperatorselects tworandompointsand
thesepositionvaluesareexchanged(Figure11). In insertion,avalue is selectedrandomlyandwillbe
insertedatanarbitraryposition(Figure12).
83
zurück zum
Buch Algorithms for Scheduling Problems"
Algorithms for Scheduling Problems
- Titel
- Algorithms for Scheduling Problems
- Autoren
- Frank Werner
- Larysa Burtseva
- Yuri Sotskov
- Herausgeber
- MDPI
- Ort
- Basel
- Datum
- 2018
- Sprache
- englisch
- Lizenz
- CC BY 4.0
- ISBN
- 978-3-03897-120-7
- Abmessungen
- 17.0 x 24.4 cm
- Seiten
- 212
- Schlagwörter
- Scheduling Problems in Logistics, Transport, Timetabling, Sports, Healthcare, Engineering, Energy Management
- Kategorien
- Informatik
- Technik