magistrsko delo Organizacija in management informacijskih sistemov
Abstract
Živimo v času, ko je na voljo velika kopica nepreglednih podatkov, ki se zbirajo preko različnih informacijskih sistemov. V okviru magistrskega dela smo se odločili analizirati podatke, ki so bili zbrani v treh letih poslovanja preko programa za vodenje poslovanja v maloprodaji (Birokrat), in skušati odkriti nove uporabne informacije, ki bi pripomogle k boljšemu poslovanju podjetja. Po preučitvi podatkov in specifično globokega prodajnega asortimenta smo se fokusirali na napovedovanje verjetnosti za prodajo posameznih izdelkov. Glavni motiv za to je bil zmanjšanje posledic napačnih odločitev. Izdelkov je veliko, napačne odločitve v nabavi pa se rezultirajo v slabi prodaji izdelkov. Podatke smo uredili in analizirali, zgradili napovedni model in nazadnje tudi uvedli napovedni model v poslovanje podjetja. Pri tem smo dosegli 50 % učinkovitejše napovedovanje prodaje v primerjavi z intuitivnim napovedovanjem poslovodje. Poleg tega smo z analizo podatkov prišli do veliko uporabnih informacij, ki jih prej ni bilo mogoče razbrati iz neobdelanih podatkov.
Keywords
odkrivanje znanja v podatkih;napovedovanje prodaje B2C;podatkovno rudarjenje;
Data
Language: |
Slovenian |
Year of publishing: |
2015 |
Typology: |
2.09 - Master's Thesis |
Organization: |
UM FOV - Faculty of Organizational Sciences |
Publisher: |
[M. Radaković] |
UDC: |
004.5 |
COBISS: |
7492627
|
Views: |
1288 |
Downloads: |
126 |
Average score: |
0 (0 votes) |
Metadata: |
|
Other data
Secondary language: |
English |
Secondary title: |
THE USE OF DATA MINING METHODS FOR PRODUCT MIX PLANNING |
Secondary abstract: |
Nowadays we live in a period of time when we obtain and collect huge amount of data through different information systems. Through our master thesis we decided to analyze and find some useful information from sales data for 3 years of sales in retail store. Data were collected with specialized information system for retail (Birokrat). Our main goal was to find useful information which would help to increase efficiency of store operations. After analyzing data and very deep product portfolio, we decided to work on predicting possibility of sales of individual products. We made this decision based on fact that product portfolio is very deep and wrong decision cause significant impact on sales. After analyzing data we made prediction model, test it and implement it within company business. Our result was 50% more precise prediction of sales of individual product in compare with old way of planning purchasing. In addition we discover a lot of useful information which will have positive impact on business in future. |
Secondary keywords: |
knowledge discovery from Data;B2C sales forecasting;data mining;business process optimization using data mining;optimization of product portfolio; |
URN: |
URN:SI:UM: |
Type (COBISS): |
Master's thesis/paper |
Thesis comment: |
Univ. Maribor, Fak. za organizacijske vede |
Pages: |
62 f. |
ID: |
9056451 |