Seite - 4 - in Algorithms for Scheduling Problems
Bild der Seite - 4 -
Text der Seite - 4 -
Algorithms 2018,11, 18
2.MILPFormulationfor theProblem
Thispaperstudiesanenergy-efïŹcientsinglemachineschedulingproblemunderTOUelectricity
tariffs. Theproblemcanalsobe calleda singlemachine schedulingproblemwith electricity costs
(SMSEC).Considerasetof jobsN={1,2, . . . ,n} thatneedtobeprocessedonasinglemachinewith
theobjectiveofminimizingtheelectricitycost. It isassumedthatall the jobsmustbeprocessedata
uniformspeed. Each job iâNhas itsuniqueprocessingtime tiandpowerconsumptionperhourpi.
Amachinecanprocess,atmost,one jobata time,andwhenit isprocessinga job,nopreemption is
allowed. Each jobandthemachineareavailable forprocessingat time instant0.Machinebreakdown
andpreventivemaintenancearenotconsidered in thispaper.
Themachine ismainlypoweredbyelectricity. Theelectricityprice followsaTOUpricingscheme
representedbyasetof timeperiodsM={1,2, . . . ,m},witheachperiodkâM,havinganelectricity
price ck andastarting timebk. The intervalofperiodk is representedby[bk,bk+1],kâM, andb1 =0 is
alwaysestablished. It isassumedthat theCmax is thegivenmakespanandbm+1â„Cmax. Thismeans
thata feasiblesolutionalwaysexists.
Themainworkof thisproblemis toassignasetof jobs toperiodswithdifferentelectricityprices
in the timehorizon [0, bm+1] tominimize total electricity cost, and themain task is todetermine to
whichperiod(s)a job isassignedandhowlonga job isprocessed ineachperiod.Hence, twodecision
variablesaregivenasfollows.Notethat thestartingtimeofeach jobcanbedeterminedbythedecision
variables (i.e.,xi,k andyi,k).
xi,k: assignedprocessingtimeof job i inperiodk, iâN,kâM;
yi,k= {
1,if job iorpartof job i isprocessed inperiodk
0,othertwise , iâN,kâM.
In addition, a job is calledprocessedwithin aperiod if both its starting timeandcompletion
timearewithin thesameperiod.Otherwise, it is calledprocessedacrossperiods [9]. Letdk andXk,
kâM, represent thedurationofperiodkandthe totalalreadyassignedprocessingtimes inperiodk,
respectively. Thedifferencebetweendk andXk isdeïŹnedas theremaining idle timeofperiodkwhich
is representedby Ik,kâM. If Ik=0, theperiodk is called full.
TheMILPmodel for thesinglemachineschedulingproblemcanbepresentedas follows:
MinTEC= n
â
i=1 m
â
k=1 pixi,kci (1)
s.t.
m
â
k=1 xi,k= ti, iâN; (2)
n
â
i=1 xi,k†bk+1âbk,kâM; (3)
xi,k†tiyi,k, iâN,kâM; (4)
lâ1
â
k=j+1 yi,kâ„ (lâ jâ1)(yi,l+yi,jâ1), iâN,3†lâ€m,1†j†lâ2; (5)
xi,kâ„ (yi,kâ1+yi,k+1â1)(bk+1âbk), iâN,2†kâ€mâ1; (6)
yi,k+yi,k+1+yj,k+yj,k+1â€3, iâN, jâN,1†kâ€mâ1, i = j. (7)
Equation(1), theobjective is tominimize the totalelectricitycost (TEC).Constraints (2)â(4)are
associatedwith theprocessingtimeassignedtoperiods.Constraints (5)and(6)areusedtoguarantee
thenon-preemptiveassumption. SpeciïŹcally, constraint (5)guarantees that if a job isprocessedacross
4
zurĂŒck 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