Raspoređivanje zasnovano na prilagodljivim pravilima

doctoral thesis

• Title (english): Scheduling based on adaptive rules
• Author: Domagoj Jakobović
• Mentor: Leo Budin (mentor)
• Committee member: Vedran Mornar (član povjerenstva)
• Committee member: Leo Budin (član povjerenstva)
• Committee member: Goran Martinović (član povjerenstva)
• Committee member: Nikola Bogunović (član povjerenstva)
• Committee member: Bojana Dalbelo-Bašić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; (Department of Electronics, Microelectronics, Computer and Intelligent Systems); Zagreb
• Defense date and country: 2005-12-07, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Universal decimal classification (UDC): 004 - Computer science and technology. Computing. Data processing

Abstract

U radu se promatra problem definiranja prikladnih postupaka raspoređivanja za različita okruženja s obzirom na uvjete raspoređivanja i zadane kriterije. Predlaže se metodologija izvođenja algoritama raspoređivanja uz pomoć genetskog programiranja. Algoritmi raspoređivanja poprimaju oblik pravila u kojima se elementima u sustavu dodjeljuje prioritet na temelju kojega se aktivnosti pridružuju sredstvima. Koristeći genetsko programiranje, izvode se pravila za različita okruženja raspoređivanja: raspoređivanje na jednom stroju, na paralelnim jednolikim strojevima, nesrodnim strojevima te u okruženju proizvoljne obrade. Za pojedino okruženje postupak raspoređivanja definiran je u dva dijela: jedan dio predstavlja meta-algoritam koji koristi prioritete elemenata u sustavu kako bi pridruživao aktivnosti sredstvima, a drugi dio predstavlja funkciju koja određuje prioritete elemenata. Prioritetna funkcija dobiva se primjenom genetskog programiranja. Za svako okruženje definirani su skupovi ispitnih primjera za učenje i ocjenu, a predloženi algoritmi uspoređeni su sa postojećim algoritmima raspoređivanja. Algoritmi raspoređivanja izvedeni uz pomoć genetskog programiranja pokazuju sličnu ili bolju učinkovitost u usporedbi s postojećim algoritmima, a značajnu prednost ostvaruju u okolinama raspoređivanja za koje ne postoje prikladni postupci raspoređivanja. U radu je također opisan postupak vrednovanja podatkovnih elemenata rješenja genetskog programiranja te postupak prilagodbe primjene genetskih operatora križanja i mutacije. Predloženi postupci prilagodbe olakšavaju pronalaženje kvalitetnog rješenja i povećavaju uspješnost evolucijskog procesa.

Abstract (english)

In this work the problem of devising an appropriate scheduling policy for different environments is addressed. The methodology which uses genetic programming to evolve scheduling heuristics is described. The scheduling heuristics are developed in the form of scheduling rules which define dynamic priorities for the elements in the system. Sheduling rules for different environments are devised using genetic programming: one machine, parallel proportional machines, unrelated machines and job shop environment. Scheduling algorithms are defined with two components: one component represents meta-algorithm which operates in scheduling environment, and the other represents an appropriate scheduling policy which derives job or machine priorities. The scheduling policy is evolved with genetic programming. For each scheduling environment a set of learning and a set of evaluation scheduling instances is defined. Devised algorithms are compared with existing algorithms in each environment. The evolved algorithms exhibit similar or better efficiency in all cases, and a significant improvement is achieved in scheduling environments where there are no fitting algorithms. Additionally, a method for evaluation of terminals in genetic programming solution and adaptive probabilities for genetic operators crossover and mutation are devised. The adaptive methods increase the probability of finding a good solution and may speed up the evolution process.

• Keywords: Raspoređivanje; genetsko programiranje; prioritetno raspoređivanje; postupci prilagodbe genetskih operatora
• Keywords (english): Scheduling heuristics; priority scheduling; genetic programming; adaptive genetic operators
• Language: croatian
• URN:NBN: urn:nbn:hr:168:435819
• Study programme: Title: Computer Science; Study programme type: university; Study level: postgraduate; Academic / professional title: Doktor znanosti (Doktor znanosti)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-07-16 07:37:24