استعمال البرمجة الديناميكية والشبكات العصبية لايجاد الخزين الامثل لمخازن الشركة العامة للزيوت النباتية == The Use of Dynamic Programming And Neural Network To Find The Optimal Stockpiling Stores General Company For Vegetable Oils
Author name:
افاق عبد الرهيب حسين محمود
Supervisor name:
فاتن فاروق صالح البدري
General topic:
Administration and Economics
Specific topic:
Operations Research
Degree:
Master
University:
University of Baghdad - Faculty Of Administration And Economics - Department Of Statistics
Language:
Arabic
University location:
Baghdad
First pages:
07T3361 - p.pdf
Abstract:
The purpose from this research to reduce the levels of inventories. in the problems of stockpiling the goals of finding the best level of stocks that are clear and accurate.We discussed this as it adopts a different input is first to focus on the style of dynamic programming in terms of properties and methods of calculations ,and method of solution using tables and the way down to find the optimal solution and for this style of algorithms. Second neural networks where this aspect to ensure a simplified study of the basic concepts of neural networks ,discussing the most important types of neural networks is the proliferation neural network and algorithms rear their own.As for the practical side, the data used are quarterly data for a period of three years (2006 - 2007 - 2008) As was initially resolved specimen using.the method of dynamic programming which were obtained on the size of inventories and return the accompanying.And in depended input dynamic programming and the results obtained by the application of dynamic programming style and application of neural networks based on the learning coefficients by trial and benefit from past experience has been obtained for less stocks possible. Down to the most important conclusions are : The decision dealt with the problem has many possibilities and where he can not resolve this problem by taking all the possibilities found style of dynamic programming to solve the problem. And the use of neural networks for stocks lower as possible
