Reševanje problema usmerjanja vozil z genetskim algoritmom

magistrsko delo

Neža Setnikar (Author), Marjetka Krajnc (Mentor)

Abstract

Zaradi večanja logističnih dejavnosti po vsem svetu je problem usmerjanja vozil eden izmed bolj znanih kombinatoričnih problemov. Splošni problem usmerjanja vozil se ukvarja z dostavo blaga strankam, za katere imamo dano njihovo povpraševanje. Rešitev predstavlja optimalna pot s čim manjšimi stroški, pri čemer moramo vse stranke obiskati natanko enkrat. Na voljo imamo več vozil, ki se začnejo in končajo v skladišču. Z večanjem velikosti problema se eksponentno povečuje kompleksnost reševanja. Zaradi tega spada problem usmerjanja vozil med NP-težke probleme, ki jih je mogoče rešiti z metahevrističnimi metodami, med katere uvrščamo tudi genetski algoritem. Magistrsko delo ima dva glavna cilja. Prvi je temeljita predstavitev problema usmerjanja vozil in genetskega algoritma. Genetski algoritem je ena izmed pomembnih tehnik za iskanje globalnega ekstrema, ki se pogosto uporablja za probleme kombinatoričnega tipa in temelji na posnemanju procesov, ki jih opazimo med naravno evolucijo. Selekcija, križanje in mutacija so glavni genetski operatorji. Drugi cilj magistrskega dela je razvoj aplikacije, ki uporabnikom omogoča rešiti problem usmerjanja vozil s pomočjo genetskega algoritma. Poleg teh dveh ciljev se v delu osredotočimo tudi na nekaj praktičnih primerov.

Keywords

problem usmerjanja vozil;genetski algoritem;križanje;mutacija;

Data

Language:	Slovenian
Year of publishing:	2019
Typology:	2.09 - Master's Thesis
Organization:	UL FMF - Faculty of Mathematics and Physics
Publisher:	[N. Setnikar]
UDC:	519.8
COBISS:	18715225
Views:	1480
Downloads:	219
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Solving vehicle routing problem using genetic algorithm
Secondary abstract:	The vehicle routing problem is one of the most known combinatorial problems due to an increase in logistics worldwide activities. The general problem is described as the delivery of goods to customers for whom their demand is given. The solution represents the optimal route with minimal transportation cost, where each customer is visited only once, by only one vehicle. Each vehicle starts and ends at the depot. The complexity of the problem increases exponentially with the size. Because of this property, the vehicle routing problem belongs to the class of NP-hard combinatorial problems that can be solved with metaheuristic methods, among which is also a genetic algorithm. This thesis has two main goals. The first is a thorough presentation of vehicle routing problem and genetic algorithm. Genetic algorithm is one of the most important global search methods commonly used for solving combinatorial problems and is based on mimicking the processes observed during natural evolution. Selection, crossover and mutation are three main genetic operators. The second goal of the master's thesis is to develop an application that allows users to solve the vehicle routing problem using a genetic algorithm. In addition to these two goals, the thesis also focuses on some practical examples.
Secondary keywords:	vehicle routing problem;genetic algorithm;crossover;mutation;
Type (COBISS):	Master's thesis/paper
Study programme:	0
Thesis comment:	Univ. v Ljubljani, Fak. za matematiko in fiziko, Oddelek za matematiko, Finančna matematika - 2. stopnja
Pages:	XI, 61 str.
ID:	11217808