magistrsko delo Organizacija in management informacijskih sistemov
Abstract
Magistrsko delo obravnava problematiko prestrukturiranja strogo hierarhične organizacijske strukture s pomočjo modelov sistemske dinamike, optimizacije z genetskimi algoritmi ter metodo končnih avtomatov. Opredeljena je metodologija definicije modela strogo hierarhične organizacijske strukture po principih sistemske dinamike. Nastavljen je sistem enačb, ki omogočajo analitično obravnavo problematike. Zaradi možnih oscilacij pri določitvi strategij prehodov je uveden končni avtomat, ki le-te preprečuje. Sistem, ki omogoča določitev strategij za doseganje želenega stanja, je razvit s pomočjo programskega jezika Javascript. Sistem je sestavljen iz simulacijskega modela, genetskih algoritmov, končnega avtomata ter uporabniškega vmesnika. Razvito rešitev je moč uporabiti prek svetovnega spleta na mobilnih napravah. Delovanje genetskih algoritmov je bilo preverjeno s pomočjo testnih funkcij. Oblikovana sta bila dva ključna scenarija za primer treh organizacijskih nivojev. Nad scenariji je bilo izvedeno 12 kombinacij nastavitev parametrov genetskih algoritmov. Podane so bile ugotovitve glede konvergence in delovanja sistema pri različnih pogojih.
Keywords
genetski algoritmi;sistemska dinamika;mobilne naprave;
Data
Language: |
Slovenian |
Year of publishing: |
2013 |
Typology: |
2.09 - Master's Thesis |
Organization: |
UM FOV - Faculty of Organizational Sciences |
Publisher: |
[B. Bavec] |
UDC: |
004.94 |
COBISS: |
7235859
|
Views: |
1410 |
Downloads: |
99 |
Average score: |
0 (0 votes) |
Metadata: |
|
Other data
Secondary language: |
English |
Secondary title: |
WEB REALIZATION OF GENETIC ALGORITHMS FOR DETERMINATION OF CONTROL STRATEGIES ON A HIERARHICAL MODEL OF HUMAN RESOURCES |
Secondary abstract: |
The thesis tackles the issue of restructuring strictly hierarhical organizational structure, using system dynamics models, genetic algorithm optimization for strategy determination and finite state automata. Metodology of strictly hierarhical organizational structure according to the principles of system dynamics is specified with a system of equations, which enables an analitical treatment of the problem. Due to the possible oscillations at determining strategies of transition through states, finite state automata is introduced. System that allows the determination of strategies for achieving the desired states is developed using the programming language JavaScript. It consists of a simulation model, genetic algotihms, finite state automata and graphical user interface. Developed solution can be used via the internet on mobile devices. Operation of genetic algorithms has been tested by using test functions. Program solution is tested through two key scenarios concerning three organizational levels. For each scenario the simulation was carried out for 12 combinations of genetic algorithm parameter settings, results and findings with regard to convergence and performance of the system were given. |
Secondary keywords: |
Genetic algorithms;Javascript;Multi-dimensional model;System dynamics;Mobile devices;Internet;Human resources;Organizational system;Control system; |
URN: |
URN:SI:UM: |
Type (COBISS): |
Master's thesis |
Thesis comment: |
Univ. v Mariboru, Fak. za organizacijske vede |
Pages: |
169 f. |
ID: |
8726997 |