20.500.12556/DKUM-94142 Odločitveni modeli za napovedovanje povpraševanja in upravljanje zalog v podjetju Tech-gum d.o.o. Decision-Making models for demand forecasting and inventory management in Tech-gum d.o.o. Magistrsko delo obravnava ključne odločitvene modele na področju napovedovanja povpraševanja in optimizacije zalog v proizvodnem podjetju Tech-gum d.o.o. Osrednji raziskovalni cilj je bil oblikovati metodološki okvir, ki bi podprl proces odločanja v proizvodnem okolju z dvojnim modelom poslovanja: izdelki za industrijsko uporabo, ki se proizvajajo po naročilu (Make-to-Order), ter potrošniški izdelki iz silikona, ki se proizvajajo na zalogo (Make-to-Stock). V teoretičnem delu so obravnavani pristopi napovedovanja, modeli planiranja proizvodnje ter metode upravljanja zalog, s posebnim poudarkom na povezavi med stabilnostjo oskrbovalne verige in finančnimi posledicami odločitev o nabavi. Posebna pozornost je namenjena dejstvu, da vsaka poslovno-ekonomsko utemeljena odločitev ni nujno tudi finančno izvedljiva, saj lahko kratkoročna likvidnost denarnih sredstev preseže pomen donosnosti, kar se še posebej odraža v gospodarskih razmerah v času pisanja dela. Analitično podlago obravnave predstavljajo modeli ekonomske količine naročila (EOQ) z razširitvami za postopno dobavo in količinske popuste, model signalne varnostne zaloge kot instrumenta obvladovanja tveganj v oskrbovalni verigi ter regresijski model napovedovanja porabe z izračunom odklonov in napak kot merila natančnosti napovedi. Takšen večdimenzionalni teoretični okvir omogoča celovito presojo o tem, kako izbrane odločitvene metode vplivajo tako na optimizacijo zalog in proizvodno učinkovitost, kot tudi na širšo finančno stabilnost podjetja. Empirični del zajema uporabo modela signalne zaloge, primerjavo različnih scenarijev oskrbe z materialom ter napovedovanje porabe surovin z uporabo metod linearnega trenda in drsečih povprečij. Rezultati napovedovanja so pokazali, da se metoda linearnega trenda izkazuje za natančnejšo od metode drsečih povprečij glede na merilo povprečne absolutne napake (MAE), kar potrjuje njeno večjo primernost za kratkoročno planiranje porabe silikona. Nadalje je bila izvedena analiza vpliva različnih modelov oskrbe na finančne kazalnike podjetja, med katerimi so ključni EBITDA, denarni tok ter razmerje med dolgom in kapitalom. Ugotovitve so pokazale, da scenarij postopne dobave kratkoročno ugodneje vpliva na likvidnost in znižuje zadolženost, medtem ko se v srednjeročnem obdobju kot ekonomsko bolj smiseln izkaže scenarij enkratne nabave, saj omogoča nižje skupne nabavne stroške in posledično višjo kapitalsko ustreznost. Hipoteze so bile zasnovane tako, da se medsebojno dopolnjujejo in tvorijo celovit analitični okvir, rezultati pa potrjujejo strateški pomen izbire ustreznih modelov, saj razlike v stroških in obsegu vezanih sredstev v zalogah pomembno vplivajo na poslovno-finančno uspešnost podjetja. Zaključki magistrskega dela potrjujejo, da je ustrezna izbira modela oskrbe in metode napovedovanja odvisna od časovnega horizonta, stabilnosti prodajnih tokov ter finančne strukture podjetja. Delo tako prispeva k boljšemu razumevanju povezave med proizvodnim planiranjem, finančnimi posledicami odločitev ter strateškim obvladovanjem zalog. The master’s thesis examines key decision-making models in the field of demand forecasting and inventory optimization in the manufacturing company Tech-gum d.o.o. The main research objective was to design a methodological framework to support the decision-making process in a production environment with a dual business model: industrial products manufactured to order (Make-to-Order) and consumer silicone products manufactured for stock (Make-to-Stock). The theoretical part discusses forecasting approaches, production planning models, and inventory management methods, with particular emphasis on the link between supply chain stability and the financial implications of procurement decisions. Special attention is devoted to the fact that not every business-economically justified decision is necessarily financially feasible, as short-term cash liquidity may outweigh profitability, which is particularly evident in the economic conditions at the time of writing this thesis. The analytical basis of the research comprises the Economic Order Quantity (EOQ) model with extensions for gradual supply and quantity discounts, the safety stock model as an instrument for risk management in the supply chain, and the regression model for consumption forecasting, supported by the calculation of deviations and errors as a measure of forecasting accuracy. Such a multidimensional theoretical framework enables a comprehensive assessment of how selected decision-making methods influence not only inventory optimization and production efficiency but also the broader financial stability of the company. The empirical part covers the application of the safety stock model, the comparison of different procurement scenarios, and the forecasting of raw material consumption using the methods of linear trend and moving averages. The forecasting results showed that the linear trend method proved to be more accurate than moving average method according to the mean absolute error (MAE), confirming its greater suitability for short-term silicone consumption planning. Furthermore, the analysis of the impact of different procurement models on the company’s financial indicators was carried out, focusing on EBITDA, cash flow, and the debt-to-equity ratio. The findings demonstrated that the gradual supply scenario has a more favorable short-term effect on liquidity and reduces indebtedness, while in the medium term the one-time purchase scenario proves to be economically more advantageous, as it enables lower overall purchasing costs and consequently a higher capital adequacy ratio. The hypotheses were designed to complement each other and form a comprehensive analytical framework, while the results confirm the strategic significance of model selection, as differences in costs and tied-up inventory levels significantly affect the company’s business and financial performance. The conclusions of the thesis confirm that the appropriate choice of procurement model and forecasting method depends on the time horizon, the stability of sales flows, and the financial structure of the company. The thesis thus contributes to a better understanding of the interconnection between production planning, the financial consequences of decisions, and the strategic management of inventories. napovedovanje povpraševanja upravljanje zalog signalna zaloga linearni trend drseče povprečje odločanje scenariji nabave finančni kazalniki MAE EOQ denarni tok demand forecasting inventory management reorder point linear trend moving average decision-making supply scenarios financial indicators MAE EOQ cash flow true true false G. Kuhta] Slovenski jezik Angleški jezik Magistrsko delo/naloga 2025-08-05 11:36:13 2025-11-03 11:13:21 2025-11-04 03:34:48 0000-00-00 00:00:00 2025 0 Maribor 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 519.8:658.7 255625731 MAG_Kuhta_Gasper_2025.pdf MAG_Kuhta_Gasper_2025.pdf 1 43ED67A6D042E3711942F29EBE8BD723 ed397165f873a385ee71ea9e31c0011441148596b2509a309b541c3ef9df80f5 f613c899-9788-11f0-88a2-00155d000105 https://dk.um.si/Dokument.php?lang=slv&id=198931 Ekonomsko-poslovna fakulteta 0 0 0