Geri Dön

Genetik algoritma ve benzetilmiş tavlama ile iki boyutlu giyotinsiz kesme problemlerine olasılıksal yaklaşım

Stochastic approach for two dimensional non-guillotineable cutting problems with genetic algorithm and simulated annealing

  1. Tez No: 135949
  2. Yazar: ALEV SÖKE
  3. Danışmanlar: YRD. DOÇ. DR. ZAFER BİNGÜL
  4. Tez Türü: Yüksek Lisans
  5. Konular: Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve Kontrol, Computer Engineering and Computer Science and Control
  6. Anahtar Kelimeler: Genetik Algoritma, Benzetilmiş Tavlama, Yerleştirme Algoritmaları, İki Boyutlu Kesme ve Paketleme Problemleri, Kırma Yaklaşımlar
  7. Yıl: 2003
  8. Dil: Türkçe
  9. Üniversite: Kocaeli Üniversitesi
  10. Enstitü: Fen Bilimleri Enstitüsü
  11. Ana Bilim Dalı: Bilgisayar Mühendisliği Ana Bilim Dalı
  12. Bilim Dalı: Belirtilmemiş.
  13. Sayfa Sayısı: 148

Özet

GENETİK ALGORİTMA ve BENZETİLMİŞ TAVLAMA ile İKİ BOYUTLU GİYOTİNSİZ KESME PROBLEMLERİNE OLASILIKSAL YAKLAŞIM Alev SOKE

Özet (Çeviri)

STOCHASTIC APPROACH FOR TWO DIMENSIONAL NON-GUILLOTINEABLE CUTTING PROBLEMS with GENETIC ALGORITHM and SIMULATED ANNEALING Alev SOKE Keywords; Genetic Algorithm, Simulated Annealing, Placement Algorithms, Two Dimensional Cutting and Packing Problems, Hybrid Approaches. ABSTRACT: Cutting problems can not be defined explicitly with a mathematical model. The solution of these problems are found by using combinational optimization in multi dimensional space. The objective of the cutting problems is to increase the usability of main object and thus to obtain the cutting pattern that has minimum trim loss values. In this thesis, A solution approach was developed for two dimensional non-guillotineable cutting problems in Matlab environment. First of all, genetic algorithms (GA) and improved bottom left (BL) algorithm were Used to solve this problems and then simulated annealing (SA) and improved BL algorithm were used. Test problems consist of different pieces: 17 and 29 regular individual rectangles to place in main object with 200x200 unit. First part of this work, an order-based GA is combined with improved BL algorithm to solve the cutting problems. This solution approach is known as hybrid GA. Firstly, hybrid GA was used to solve 1 7 pieces test problems. The influences of the different population sizes and mutation rates on the solution of these problems were examined. Base on this examination, it is observed that as the population size increases the trim loss value decreases. There is very small effect of mutation rate on the solution. Secondly hybrid GA was applied to 29 pieces a test problem. The influences of different crossover techniques on the solution of the cutting problem were studied. At the result of this study, it is seen that the best result is obtained with order based crossover technique. Second part of this work, SA and improved BL algorithm are combined to solve the same cutting problems. This solution approach is known as hybrid SA. Firstly, hybrid SA was applied to solve the 17 pieces test problems.The influences of different cooling schemes, neighbourhood moves and values for equilibrium condition on the solution of the cutting problems were examined for the different temperature values. Secondly, the solution for 29 pieces a test problem was solved using parameteres of the best results obtained in previous work. At the result of these simulations, the trim loss values of the 17 and 29 pieces test problems obtained by using hybrid GA are varied between 2% and 11%, 5% and 9% respectively. Similarly, the trim loss values of the 17 and 29 pieces test problems obtained by using hybrid SA are varied between 4% and 21%, 10% and 17% respectively. Finally, it is seen that hybrid GA gives better results than hybrid SA for cutting problems. in

Benzer Tezler

  1. Manipulation and placement planning for loading a dishwasher by a robot

    Bulaşık makinesini robotla doldurmak için manipulasyon ve yerleştirme planlaması

    BAHAR İRFAN

    Yüksek Lisans

    İngilizce

    İngilizce

    2016

    Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolBoğaziçi Üniversitesi

    Bilgisayar Mühendisliği Ana Bilim Dalı

    PROF. DR. HÜSEYİN LEVENT AKIN

  2. Yapay zeka tekniklerine dayanan yöntemlerle çok sayıda hedef izleme

    Multiple target tracking with the methods based on artificial intelligence techniques

    İLKE TÜRKMEN

    Doktora

    Türkçe

    Türkçe

    2005

    Elektrik ve Elektronik MühendisliğiErciyes Üniversitesi

    Elektronik Mühendisliği Ana Bilim Dalı

    PROF.DR. KERİM GÜNEY

  3. Kesme ve yükleme problemleri için lagrange fonksiyonları temelli çözüm yöntemleri

    Lagrange function based solution methods for cutting and packing problems

    BANU İÇMEN ERDEM

    Doktora

    Türkçe

    Türkçe

    2021

    Endüstri ve Endüstri MühendisliğiEskişehir Teknik Üniversitesi

    Endüstri Mühendisliği Ana Bilim Dalı

    PROF. DR. REFAİL KASIMBEYLİ

  4. Hipersezgisel yöntemlerle lojistik ağ tasarımı ve optimizasyon

    Logistic network design and optimization using hyperheuristic methods

    VURAL EROL

    Doktora

    Türkçe

    Türkçe

    2017

    Endüstri ve Endüstri Mühendisliğiİstanbul Teknik Üniversitesi

    Endüstri Mühendisliği Ana Bilim Dalı

    DOÇ. DR. MURAT BASKAK

    PROF. DR. GÜLGÜN KAYAKUTLU

  5. İki boyutlu kesme ve ana malzeme seçimi problemleri için matematiksel modeller ve çözüm yaklaşımları

    Mathematical models and solution approaches for two dimensional cutting and assortment problems

    BANU İÇMEN

    Yüksek Lisans

    Türkçe

    Türkçe

    2015

    Endüstri ve Endüstri MühendisliğiAnadolu Üniversitesi

    Endüstri Mühendisliği Ana Bilim Dalı

    PROF. DR. REFAİL KASIMBEYLİ