doctoral thesis
Managing critical success factors of ERP implementation by using Bayesian probability networks

Milorad Nikitović (2014)
Sveučilište u Zagrebu
Fakultet organizacije i informatike Varaždin
Metadata
TitleUpravljanje kritičnim čimbenicima uspješnosti primjene ERP sustava pomoću Bayesovih mreža vjerojatnosti
AuthorMilorad Nikitović
Mentor(s)Vjeran Strahonja
Abstract
Cilj rada je istraživanjem utvrditi što potpuniji skup utjecajnih kritičnih čimbenika uspjeha primjene ERP (eng. Enterprise Resource Planning) sustava koji će poslužiti kao podloga za izradu modela upravljanja primjene ERP sustava pomoću Bayesove mreže. Ostvarivanje ovog cilja omogućiti će izradu smjernica za uvećanje ukupne uspješnosti primjene ERP rješenja na temelju određenog skupa kritičnih čimbenika. Težinskim faktorima su utvrđene vrijednosti značaja izabrana 32 kritična čimbenika uspješnosti za cjelovitost ERP primjene. Istovremeno je istražen utjecaj istih kritičnih čimbenika na šest faza životnog ciklusa pri čemu su uočene razlike, što sugerira da je proces primjene ERP-a bolje voditi prema fazama životnog ciklusa. Izrađeni su modeli uspješnosti primjeneERP rješenja sukladno izabranim kritičnim čimbenicima a u okviru sedam grupa čimbenika.Modeli predstavljaju široku mogućnost predviđanja, sugeriranja i praćenja nivoa kritičnih čimbenika uspješnosti koji sudjeluju u procesu primjene. U svrhu upravljanja primjenom je izrađeno devet Bayesovih mreža. Simulacija Bayesovom mrežom ukazuje kako kritični čimbenici uspješnosti utječu na grupu čimbenika kojoj pripadaju, na ukupnost procesa primjene ali i na pojedinačne čimbenike povezane od strane algoritma koji je primijenjen u definiranju mreže.Tijekom provođenja empirijskog istraživanja upotrijebljeni su analitički alati MATLAB, Statistics, GeNie&Smile, Hugin, Netica, Orange Canvas, programski jezik C++ te Pervasive SQL.Ukupnost pristupa se temeljina analizi selekcije varijabli odnosno kritičnih čimbenika, prikupljanjem podataka, njihovim vrednovanjem, te definiranjem modela kritičnih čimbenika kao varijabli modela Bayesovih mreže. Analiza modela se temeljila na prikazu utjecaja svakog pojedinačnog čimbenika na ukupnost procesa, grupu ili druge čimbenike.Simulacijom Bayesovom mrežom je utvrđeno da iste predstavljaju dobar alat za ocjenu kako značaja pojedinačnog čimbenika na ukupnost procesa, grupu čimbenika kojoj pripada kao i na ostale čimbenike predstavljene mrežom.
KeywordsERP critical success factors ERP implementation life cycle phases Bayesian network
Parallel title (English)Managing critical success factors of ERP implementation by using Bayesian probability networks
Committee MembersDiana Šimić (committee chairperson)
Vjeran Strahonja (committee member)
Katarina Ćurko (committee member)
GranterSveučilište u Zagrebu
Fakultet organizacije i informatike Varaždin
PlaceVaraždin
StateCroatia
Scientific field, discipline, subdisciplineSOCIAL SCIENCES
Information and Communication Sciences
Information Systems and Information Science
UDK005
GENERALLY
Management
Study programme typeuniversity
Study levelpostgraduate
Study programmePostgraduate doctoral study in Information Science
Academic title abbreviationdr.sc.
Genredoctoral thesis
Language Croatian
Defense date2014-10-16
Parallel abstract (English)
The objective of the dissertation is to use research to determine the most complete set of influential critical success factors in the implementation of ERP systems as possible, which will then serve as the basis for designing an ERP system implementation management model using the Bayesian network. The achievement of this objective will enable the development of guidelines for the enhancement of the overall success of ERP solution implementation on the basis of a determined set of critical factors.The significance values of the chosen 32 critical success factors for ERP implementation completeness have been determined by weighting factors. Simultaneously, the impact of the same critical factorson six life cycle phases has been explored. Differences have been noted which suggest that it is better to manage the process of ERP implementation according to the life cycle phases.The models of ERP solution implementation success has been designed inaccordance with the chosen critical factors, within seven factor groups.The models represents a broad possibility of predicting, suggesting and monitoring levels of critical success factors which take part in the implementation process. Nine Bayesian networks have been designed with the purpose of managing the implementation. The Bayesian network simulation shows how the critical success factors influence the group of factors to which they belong, the completeness of the implementation process, but also the single factors which are linked through the algorithm used in defining the network.During the empirical research, the following analytical tools were used: MATLAB, Statistics, GeNie&Smile, Hugin, Netica, C++ programming language and Pervasive SQL.The completeness of the approach was based on the analysis of the selection of variables i.e. critical factors, with data acquisition and assessment, and also with defining the model of critical factors as variables of the Bayesian network. The analysis of the model was based on the presentation of the influence of each single factor on the completeness of the process, the group or other factors.The Bayesian network simulation determined that this network represents a good evaluation tool for the significanceof a single factor to the completeness of the process, the group of factors it belongs to and the other factors presented in the network.
Parallel keywords (Croatian)ERP kritični čimbenici uspješnosti faze životnog ciklusa primjene ERP-a Bayesova mreža
Versionaccepted version
Resource typetext
Access conditionOpen access
Terms of usehttp://rightsstatements.org/vocab/InC/1.0/
Noteaccepted version
URN:NBNhttps://urn.nsk.hr/urn:nbn:hr:211:753585
CommitterLjiljana Hajdin