20.500.12556/DKUM-19119 REŠEVANJE NELINEARNEGA LOKACIJSKEGA PROBLEMA Z UPORABO NUMERIČNIH OPTIMIZACIJSKIH METOD NA PRIMERU REALNIH PODATKOV DISTRIBUCIJSKE ORGANIZACIJE SOLVING NON - LINEAR LOCATION PROBLEM WITH USE OF NUMERICAL OPTIMISATION METHOD, AN EXAMPLE OF REAL DISTRIBUTION INFORMATION Diplomsko delo prikazuje uporabo nelinearnih multidimenzionalnih metod na primeru realnih podatkov distribucijske organizacije. V delu želimo prikazati način, kako določiti geografsko lokacijo nekega centralnega skladišča za potrebe dnevne distribucije blaga široke potrošnje. Ker gre za nelinearni multidimenzionalni optimizacijski problem, je potrebno za določitev optimalne lokacije uporabiti nelinearne multidimenzionalne optimizacijske metode. V delu je najprej predstavljenih nekaj najbolj tipičnih tovrstnih metod, sama rešitev problema pa je izpeljana s pomočjo naslednjih dveh metod: Hooke-Jeevesove metode vzorčnih premikov in Nelder-Meadove simpleksne metode. V delu je podan prikaz vseh izračunanih numeričnih rezultatov obeh metod za optimalno lokacijo skladišča in predlagano implementacijo dobljene rešitve. This study shows the use of non-linear multidimensional methods in everyday life. In this work, we wish to show how to set a geographical location of the given central warehouse for needs of daily distribution of goods for general consumers. The goal of this study is to help the reader when solving similar problems. Everyone who will read the study will be able to decide based on their own assessment, which methods are appropriate for their examples. He or she will understand how to find the most appropriate location, which method to use and how to imply it in the daily life. In study we first set the problem and the objective that we are looking for. We define variables and restrictions. For a better explanation of the variables, we have used some simplifications, which are explained in the study. The first part shows us one of the world-wide mostly used methods. In the second part we have shown calculation procedure two different models of optimization for the given problem. For the final part gives a comparison and analysis of the methods for solving location problems Optimizacija numerične optimizacijske metode nelinearno programiranje Hooke-Jeevesova metoda vzorčnih premikov Nelder-Meadova simpleksna metoda Cauchyeva metoda najstrmejšega spusta Newtonova multidimenzionalna metoda Marquardtova metoda reševanje lokacijskih problemov Optimization numerical optimization method nonlinear programming Hooke-Jeeves pattern search method Nelder-Mead simplex method Cauchy’s steepest descent method Newton’s multidimensional method Marquardt’s method true true false Slovenski jezik Angleški jezik Diplomsko delo 2011-06-21 14:10:42 2011-10-16 13:42:53 2022-04-13 15:06:18 2023-12-25 03:10:14 2011 0 Maribor 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 URN:SI:UM:DK:XBDZWRIH UNI_Korosec_Miroslav_2011.pdf UNI_Korosec_Miroslav_2011.pdf 1 2EB72C30005CBAB1C8A127B92C6C6B55 fb1d2bb17f44d0840fa799d7b06c03f6847377fea96c991b3da11693d8728ce8 4be010a4-7c05-11eb-bb7a-00155d0001ca 20.500.12556/dkum/6d527980-b420-44da-839d-2c70f78edff5 https://dk.um.si/Dokument.php?lang=slv&id=22987 Fakulteta za logistiko 0 0 0