Analiza odabranih metoda dubinske obrade podataka na primjeru lanaca nabave

master's thesis

• Title (english): Analysis of selected data mining methods applied to supply chain data
• Author: Aleksandar Grabant
• Mentor: Nikola Vlahović (mentor)
• Committee member: Mirjana Pejić Bach (član povjerenstva)
• Committee member: Blaženka Knežević (član povjerenstva)
• Granter: University of Zagreb; Faculty of Economics and Business; Zagreb
• Defense date and country: 2023-09-26, Croatia
• Scientific / art field, discipline and subdiscipline: SOCIAL SCIENCES; Economics; Business Informatics

Abstract

Umjetna inteligencija je već sada postala neizostavni dio mnogih multinacionalnih kompanija. One su takvu tehnologiju prepoznale na vrijeme te ju koriste kako bi svoje poslovne modele i procese unaprijedile na novu razinu. Jedan od takvih primjera je Amazon koji koristi umjetnu inteligenciju u svojim pametnim trgovinama. Potreba za dubinskom obradom podataka se javila upravo iz razloga što poduzeća svakodnevno generiraju velike količine podataka, a u tim podacima se često kriju korisne informacije koje je potrebno otkriti te iz njih stvoriti nova znanja. Osim u poslovnom svijetu, umjetna inteligencija se koristi i u medicini, javnoj upravi, prometu te drugim područjima.
Cilj istraživanja je istražiti korisnost te načine primjene umjetne inteligencije u lancima nabave i usporediti softvere te generirane rezultate na temelju metoda dubinske obrade podataka.
Jedno od područja u kojima se intenzivno koristi umjetna inteligencija je i poslovna funkcija nabave odnosno upravljanje lancima nabave. Cilj upravljanja lancima nabave je optimizirati sam poslovni proces, a jedan od načina je i primjena umjetne inteligencije. Na primjeru podataka iz proizvodnog poduzeća, u ovom radu, prikazana su dva softvera za dubinsku obradu podataka, Weka i Statistica te su korištene tri metode dubinske obrade podataka. Riječ je o umjetnim neuronskim mrežama, metodi slučajnih šuma te stablu odlučivanja.
Rezultati su pokazali da poduzeća imaju veliku korist od primjene metoda dubinske obrade podataka te su odabrane metode dubinske obrade podataka s visokom preciznošću predvidjele rezultate o prihvatljivosti, odnosno ne prihvatljivosti pojedinog dobavljača.

Abstract (english)

Artificial intelligence has already become an indispensable part of many multinational companies. They recognized such technology in time and use it to advance their business models and processes to a new level. One example is Amazon, which uses artificial intelligence in its smart stores. The need for in-depth data processing arose precisely because companies generate large amounts of data every day, and this data often hides useful information that needs to be discovered and new knowledge created from it. Apart from the business world, artificial intelligence is also used in medicine, public administration, transport and other fields.
The goal of the research is to investigate the usefulness and ways of applying artificial intelligence in supply chains and to compare software and generated results based on in-depth data processing methods.
One of the areas in which artificial intelligence is intensively used is the business function of procurement, i.e. supply chain management. The goal of supply chain management is to optimize the business process itself, and one of the ways to do it is the application of artificial intelligence. On the example of data from a manufacturing company, in this paper, two software for in-depth data processing, Weka and Statistica, are presented, and three methods of in-depth data processing are used. The methods that were used are artificial neural networks, random forest and decision tree.
The results showed that companies benefit greatly from the application of in-depth data processing methods, and the selected in-depth data processing methods predicted with high precision the results about the acceptability or non-acceptability of an individual supplier.

• Keywords: umjetna inteligencija; umjetna neuronska mreža; stablo odlučivanja; slučajne šume; lanac nabave
• Keywords (english): artificial intelligence; artificial neural network; decision tree; random forests; supply chain
• Language: croatian
• URN:NBN: urn:nbn:hr:148:027807
• Study programme: Title: Business Economics; Study programme type: university; Study level: integrated undergraduate and graduate; Academic / professional title: magistar/magistra ekonomije (magistar/magistra ekonomije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Access restricted to students and staff of home institution
• Created on: 2023-09-26 18:40:19