Page - 45 - in Algorithms for Scheduling Problems
Image of the Page - 45 -
Text of the Page - 45 -
Algorithms 2018,11, 80
To show that Problem 2 is stronglyNP-hard, we prove that its decision version is strongly
NP-complete. Let M be some threshold value for the makespan. Without loss of generality,
weassumethat this threshold isattainable, i.e., there isa feasibleschedulewith theobjectivevalueM.
Then, thedecisionversionofProblem2canbeformulatedas follows:
Problem2-D.Foran instanceofproblem1|rj|Lmax (the input), amongall feasible scheduleswitha
givenmaximumjob lateness,does thereexistonewith themakespannotexceedingmagnitudeM
(here, theoutput isa“yes”ora“no”answer)?
Theorem1. Problem2-Dis stronglyNP-complete.
Proof. Tobeginwith, it is easy tosee that theproblemis inclassNP.First,weobserve that thesize
of an instance of problem1|rj|Lmax with n jobs isO(n). Indeed, every job jhas threeparameters,
rj,pj and dj; letMbe themaximumsuchmagnitude, i.e., themaximumamongall jobparameters.
Then, the number of bits to represent ourproblem instance is boundedaboveby 3n logM. Since
constant M fitswithin a computerword, the size of the problem isO(n). Now, given a feasible
schedulewithsomemaximumjoblateness,wecanclearlyverify itsmakespaninnomorethann steps,
which isapolynomial in thesizeof the input.Hence,Problem2-Dbelongs toclassNP.
Next,weuse the reduction fromastronglyNP-complete three-PARTITIONproblemto show
theNP-hardnessofourdecisionproblem. In three-PARTITION,wearegivenasetAof3melements,
their sizesC={c1,c2, . . . ,c3m}andanintegernumberBwith 3m
∑
i=1 ci=mB,whereas thesizeofevery
element fromsetA isbetweenB/4andB/2.Weareaskedif thereexists thepartitionofsetA intom
(disjoint) setsA1, . . . ,Am suchthat thesizeof theelements ineverysubset sumsuptoB.
Given an arbitrary instance of three-PARTITION,we construct our scheduling instancewith
3m+m jobswith the total lengthof n
∑
ι=1 cι+mas follows.Wehave3mpartition jobs1,. . . ,3mwith
pi= ci, ri=0anddi=2mB+m, for i=1,. . . ,3m. Besides,wehavem separator jobs j1, . . . , jmwith
pjι = 1, rjι = Bι+ ι−1, djι = Bι+ ι. Note that this transformation creatinga scheduling instance
ispolynomialas thenumberof jobs isboundedbythepolynomial inm, andallmagnitudescanbe
represented inbinaryencoding inO(m)bits.
First,weeasilyobserve that thereexist feasible schedules inwhich themaximumjob lateness
is zero:wecan just schedule theseparator jobsat their release timesand include thepartition jobs
arbitrarilywithoutpushinganyseparator job. Now,wewish toknowwhetheramongthe feasible
scheduleswith themaximumjob latenessofzero there isonewithmakespanmB+m.Weclaimthat
thisdecisionproblemhasa“yes”answer iff thereexistsasolutionto three-PARTITION.
Inonedirection, suppose three-PARTITIONhasasolution. Then,weschedule thepartition jobs
corresponding to the elements fromsetA1within the interval [0,B] andpartition jobs fromsetAi
within the interval [(i−1)B+ i−1, iB+ i−1], for i=2,. . . ,m (weschedule the jobs fromeachgroup
inanon-delay fashion,using, for instance, theEDheuristics).Weschedule theseparator job jiwithin
the interval [iB+ i−1, iB+ i], for i=1,. . . ,m. Therefore, the latest separator jobwillbecompletedat
timemB+m, andthemakespanin theresultantschedule ismB+m.
Intheotherdirection,supposethereexistsafeasiblescheduleSwithmakespanmB+m. Sincethe
total sumof jobprocessingtimes ismB+m, theremayexistnogapinthatschedule.However, then,
the time intervalof lengthBbeforeeveryseparator jobmustbefilled incompletelywith thepartition
jobs,whichobviouslygivesasolutionto three-PARTITION.
3.BasicDefinitionsandProperties
Thissectioncontainsnecessarypreliminaries to theproposedframework includingsomeuseful
properties of the feasible schedules created by the ED heuristics (we call them ED-schedules).
Our approach is basedon the analysis of theproperties ofED-schedules. This analysis is initially
carriedout for thedecisionversionof single-criterionproblem1|rj|Lmax: we shall particularly be
45
back to the
book Algorithms for Scheduling Problems"
Algorithms for Scheduling Problems
- Title
- Algorithms for Scheduling Problems
- Authors
- Frank Werner
- Larysa Burtseva
- Yuri Sotskov
- Editor
- MDPI
- Location
- Basel
- Date
- 2018
- Language
- English
- License
- CC BY 4.0
- ISBN
- 978-3-03897-120-7
- Size
- 17.0 x 24.4 cm
- Pages
- 212
- Keywords
- Scheduling Problems in Logistics, Transport, Timetabling, Sports, Healthcare, Engineering, Energy Management
- Categories
- Informatik
- Technik