Fakülteler / Faculties

Permanent URI for this communityhttps://hdl.handle.net/11727/1395

Browse

Filters

2006 - 20092

Settings

Subject: search.filters.subject.simulation optimization

Search Results

Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    Simulation Optimization Based DSS Application: A Diamond Tool Production Line in İndustry
    (2006) Dengiz, B.; Bektas, T; Ultanir, AE; 0000-0003-0634-144X
    A diamond tool manufacturing system simulation is developed to predict the number of machines and the number of workers necessary to maintain desired levels of production for a company in Ankara, Turkey. The current manufacturing system is analysed by a simulation model emphasizing the bottlenecks and the poorly utilized machines. Validated simulation outputs are collected and used to build a multiple regression meta-model as a simulation optimization based decision support system (DSS). The proposed DSS involves analysis and evaluation of the system's behaviour through the use of a meta-model with an integrated optimization module. It enables the decision maker to perform sensitivity analysis by considering several combinations of decision variables. The aim of this study is two fold. The first is to represent a simulation optimization based DSS application for a real system by considering all the required steps. The second is to analyse the performance of the current production system and determine the optimum working conditions by simulation with greatly reduced cost, time, and effort. (c) 2005 Elsevier B.V. All rights reserved.
  • No Thumbnail Available
    Item
    Optimization of Manufacturing Systems Using A Neural Network Metamodel with A New Training Approach
    (2009) Dengiz, B.; Alabas-Uslu, C.; Dengiz, O.
    In this study, two manufacturing systems, a kanban-controlled system and a multi-stage, multi-server production line in a diamond tool production system, are optimized utilizing neural network metamodels (tst_NNM) trained via tabu search (TS) which was developed previously by the authors. The most widely used training algorithm for neural networks has been back propagation which is based on a gradient technique that requires significant computational effort. To deal with the major shortcomings of back propagation (BP) such as the tendency to converge to a local optimal and a slow convergence rate, the TS metaheuristic method is used for the training of artificial neural networks to improve the performance of the metamodelling approach. The metamodels are analysed based on their ability to predict simulation results versus traditional neural network metamodels that have been trained by BP algorithm (bp NNM). Computational results show that tst NNM is superior to bp NNM for both of the manufacturing systems. Journal of the Operational Research Society (2009) 60, 1191-1197. doi:10.1057/palgrave.jors.2602620 Published online 30 July 2008