Publication detail
Stochastic Heuristics for Knapsack Problems
ŠEDA, M. ŠEDA, P.
Czech title
Stochastické heuristiky pro problémy batohu
English title
Stochastic Heuristics for Knapsack Problems
Type
journal article in Scopus
Language
en
Original abstract
In this paper, we introduce knapsack problem formulations, discuss their time complexity and propose their representation and solution based on the instance size. First, deterministic methods are briefly summarized. They can be applied to small-size tasks with a single constraint. However, because of NP-completeness of the problem, more complex problem instances must be solved by means of heuristic techniques to achieve an approximation of the exact solution in a reasonable amount of time. The problem representations and parameter settings for a genetic algorithm and simulated annealing frameworks are shown.
Czech abstract
V tomto příspěvku jsou představeny formulace verzí problému batohu, je diskutována jejich časová složitost a navrženo řešeno v závislosti na velikosti instance. V úvodu jsou stručně shrnuty deterministické metody. Ty mohou být aplikovány na úlohy malého rozsahu s jením omezením. Avšak vzhledem k NP-úplnosti problému instance většího rozsahu je nutné řešit pomocí heuristických metod, aby bylo možné získat aproximaci řešení v dostupném čase. V příspěvku je prezentována reprezentace problémů a nastavení parametrů pro genetický algoritmus a simulované žíhání.
English abstract
In this paper, we introduce knapsack problem formulations, discuss their time complexity and propose their representation and solution based on the instance size. First, deterministic methods are briefly summarized. They can be applied to small-size tasks with a single constraint. However, because of NP-completeness of the problem, more complex problem instances must be solved by means of heuristic techniques to achieve an approximation of the exact solution in a reasonable amount of time. The problem representations and parameter settings for a genetic algorithm and simulated annealing frameworks are shown.
Keywords in Czech
problém batohu, dynamické programování, metoda větví a mezí, genetický algoritmus, simulované žéhání
Keywords in English
knapsack problem, dynamic programming, branch and bound method, heuristic, genetic algorithm, simulated annealing
Released
05.08.2018
Publisher
Springer-Verlag
Location
Berlin
ISSN
2194-5357
Volume
837
Number
1
Pages from–to
157–166
Pages count
10
BIBTEX
@article{BUT149171,
author="Miloš {Šeda} and Pavel {Šeda},
title="Stochastic Heuristics for Knapsack Problems",
year="2018",
volume="837",
number="1",
month="August",
pages="157--166",
publisher="Springer-Verlag",
address="Berlin",
issn="2194-5357"
}