Geri Dön

A multi-agent systems based conceptual ship design decision support system

Başlık çevirisi mevcut değil.

  1. Tez No: 402716
  2. Yazar: BEKİR SITKI TÜRKMEN
  3. Danışmanlar: DR. OSMAN TURAN
  4. Tez Türü: Doktora
  5. Konular: Deniz Bilimleri, Denizcilik, Gemi Mühendisliği, Marine Science, Marine, Marine Engineering
  6. Anahtar Kelimeler: Belirtilmemiş.
  7. Yıl: 2005
  8. Dil: İngilizce
  9. Üniversite: University of Strathclyde
  10. Enstitü: Yurtdışı Enstitü
  11. Ana Bilim Dalı: Belirtilmemiş.
  12. Bilim Dalı: Belirtilmemiş.
  13. Sayfa Sayısı: 247

Özet

Özet yok.

Özet (Çeviri)

In this study, the integration of multi-criteria decision-making and performance evaluation tools into a flexible and distributed ship design decision support system is proposed. In order to achieve such a distributed support system, an agent architecture, which is based on encapsulation of expertise, conflict resolution strategies, learning, and performance evaluation concepts, is developed. A multiagent system is also devised in this study by utilising the agent architecture and further exploiting the research towards multi-criteria decision-making and multiagent negotiation for the integration. The proposed multi-agent system is applied to multi-objective optimisation and internal hull subdivision problem by concentrating on the negotiation aspects of the agents. The results of the case studies show that the developed methodologies offer improvements to the performance of designs, reduction in the calculation cost, and reuse of knowledge in conflict resolution. A priori preference introduction method is also developed to improve the decision maker's involvement to the multi-objective optimisation problem in this thesis. The developed method is extensively tested with both numerical functions and an internal hull subdivision problem of a ROPAX vessel. The proposed method can effectively address the guidance for convex, non-convex and real world problems. A new solution methodology with changes to the genetic coding and operators for the internal hull subdivision optimisation problem is proposed and successfully implemented. The application of new solution methodology demonstrates that the number of bulkheads for the internal hull subdivision problem can easily be determined in one optimisation run. Recommendations for future research are given in this thesis as follows: An expert system approach for conflict resolution, performance evaluation and learning, an interactive virtual environment by the integration of proposed multi-agent system into a 3D real time simulation environment, and finally improvements to the learning layer of the agent architecture.

Benzer Tezler

  1. Kişiselleştirilmiş öğrenmeyi destekleyen bir altyapı tasarımı

    The design of an infrastructure supporting personalized learning

    BİROL ÇİLOĞLUGİL

    Doktora

    Türkçe

    Türkçe

    2015

    Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolEge Üniversitesi

    Bilgisayar Mühendisliği Ana Bilim Dalı

    PROF. DR. MUSTAFA MURAT İNCEOĞLU

  2. Avangart tasarım süreci için kent morfolojisi kısıtlayıcıları ile bir simülasyon modeli

    A simulation model with urban morphology constraints for avant-garde design process

    ASLI AĞIRBAŞ

    Doktora

    Türkçe

    Türkçe

    2016

    MimarlıkMimar Sinan Güzel Sanatlar Üniversitesi

    Mimarlık Ana Bilim Dalı

    DOÇ. DR. FETHİYE EMEL ARDAMAN

  3. A multi-agent adaptive learning system for distance education

    Uzaktan öğrenmeye yönelik adapte olabilen çok aracılı sistem

    FATMA CEMİLE SERÇE

    Doktora

    İngilizce

    İngilizce

    2008

    Bilgisayar Mühendisliği Bilimleri-Bilgisayar ve KontrolOrta Doğu Teknik Üniversitesi

    Bilişim Sistemleri Ana Bilim Dalı

    DOÇ. DR. FERDA NUR ALPASLAN