Genetic algorithms using multi-objectives(12)

We are interested in a job-shop scheduling problem corresponding to an industrial problem. Gantt diagram’s optimization can be considered as an NP-difficult problem. Determining an optimal solution is almost impossible, but trying to improve the current s

186A.Cardonetal./RoboticsandAutonomousSystems33(2000)179–190

8.2.Thedifferenttypesofagents

Thelocaldatabasecontainsthebasicknowledgeoftheassociatednodesandalsoinformationonthepro-gressionstateofthecooperativenegotiationaswellasthestateoftheresolutionprocess[57].Thetaskofthecommunicationmanageristoestablishcommu-nicationswithotheragents.Itistheuniquecompo-nentofanodethatisindirectrelationshipwiththesystem.Especially,itensuresthereceptionandsendsmessages.

Thecontractmanager’sjobistoexaminethe“auctioned”task,thecompliancewiththecontractanditsending.Inotherwords,thecontractmanagerensuresthecoordinationofallagents[14].Thetaskadministratorisresponsiblefortheprogressmadeinaprocessandtheresultsofataskassignedtoagivenagent.Itreceivestheproblemthatneedstobesolvedfromthecontractmanager.Itusesthelocaldatabaseinorderto ndasolutionandgivesitbacktothecontractmanager.

Thejobofacontract-netbasedsystemoftenbeginswithaproblemdivisionstage[15].Afterthat,theprob-lemtobesolvedisdividedintoasetofsub-problems[25].Aspecialagent,themanager,assignstaskstoasub-problem.Themanagerissuesapublicoffer,calledacontractforeachsub-problemtobesolvedaccord-ingtotheschemede nedbySmith[48].

Becauseofthedistributednatureoftheproblem,wechoseanagentbasedmodelization,asimpli edversionofFISIAS[41].Thedifferentelementsoftheenvironment(theuniverse)canbethemachines(andgroupsofmachines),thejobs(andgroupsofjobs),theGanttdiagram(anattribute)anddistributoragentofjobs(DAJ)(Fig.4).Universeelementsarethe“objects”thatcanreceiveorsenddatatoagents.Machinescanbeseenasagentswhosetaskistoperformthejoboperationatagiventime.However,themachinecanalsobeseenasapurelyreactiveagent,which,accordingtothejob,canreply:“Icandoitornot”(Fig.5).

Then,wecanusethecontractnet(Fig.6)de nedbySmith[48],whosemanager,theDAJ,willproposetheallocationofajobtoamachinethroughtheuseofanegotiationagentdelegatedforthisjob(NA).Ac-cordingtotheinformationgivenbyamachine,theNAwillestablishacontractbetweenamachineMiandtheDAJ[5].ButtheDAJcanalwaysproposeajob

to

Fig.4.Representationoftheuniverse.

severalmachinestocreatesomecompetitionbetweenmachineagents.However,theDAJcanbreakthiscon-tractatanytimeiftheagentisunabletocomplywiththe

contract.

Fig.5.Firstrepresentationofthecontractnet.

搜索“diyifanwen.net”或“第一范文网”即可找到本站免费阅读全部范文。收藏本站方便下次阅读，第一范文网，提供最新教学研究Genetic algorithms using multi-objectives(12)全文阅读和word下载服务。