الخوارزمية الجينية في السيطرة على الخزين الانتاجي المتعدد : بحث تطبيقي == Genetic Algorithm For Controlling Multi - Production Inventory (Applied Research)
Author name:
الاء حكمة عبد الستار البياتي
Supervisor name:
عبد المنعم كاظم حمادي الشكري
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:
07T3422 - p.pdf
Abstract:
تضمن هذا البحث حل لمشكلة السيطرة على الخزين الانتاجي المتعدد اعتمادا على نماذج الخزين المحددة للمنشات الانتاجية في حالة وجود وعدم وجود عجز لبعض منتجات شركة مصافي الوسط الاكثر طلبا في اسواقنا المحلية، اذ تم اختيار سبع مواد للدراسة وهي (البنزين، النفط الا | This research dealt with solving the problem of control of inventories multi - productive based on determenestic inventeries models for production institutions with and without shortage for the most requested productions in our local markets for Midland Refineries Company, such that seven materials have been selected of the study, namely, (Benzene, White Oil, Jet Fuel, Gas Oil, Grease, Oxidized Asphalt and Liquid Gas).Firstly the traditional methods have been applied in extraction the optimal economic quantities and the total cost of the inventoies, for the purpose of improving the results, the genetic algorithm has been used on the results obtained from traditional methods, which is one of artificial intelligence algorithms that rely on the mechanism of random search and selection of the science of natural genetics. Roulette wheel selection has been used as a method to select a set of chromosomes and one crossover point to generate new society, and the mutation has been applied by change only one bit from any chromosome gas been selected randomlly to increase diversity in the new society and try to solve the optimization solution according to the value of the fitness function. The results of applying traditional methods in inventories models showed shortage cases in production relative to demand for materials (Benzene, White Oil, Jet Fuel, Gas Oil, Grease, Oxidized Asphalt and Liquid Gas) proportion rates ranging between (194.63% - 722.46%) and an increase in production up to 98% in Oxidized Asphalt material.The use of genetic algorithm led to reduce the shortage in production relative to demand for Benzene, White Oil, Jet Fuel, Gas Oil and Liquid Gas at proportion rates ranging between (21.64% - 75.92%). Also the results showed that the use of genetic algorithm contributed to cover the shortage in production as compared with the traditional methods at change rates (14.01% - 28.61%) for Benzene material, (13%) for Jet Fuel material and (23%) for Liquid Gas material. The results of genetic algorithm improved the results of the traditional methods for Benzene, White Oil, Jet Fuel, Gas Oil, Grease, Oxidized Asphalt and Liquid Gas materials at rates ranging between (12.98% - 42.66%). And finally the use of genetic algorithm reduced the total cost of the inventories of all the company's products at rate 26.15%.This thesis contained four chapters : the first chapter includes : introduction, the goal of the research, the research problem and refrence review. The second chapter includes : the theoretical side distributed over two sections first topic the basic concepts of the system of inventories; the second section includes an explanation of the style of a genetic algorithm and method of implementation. Third chapter includes : the practical side (display data that has been assessed from Midland Refineries Company distributed throughout the whole year, extract the optimal economic quantities by traditional methods and improve them by using the genetic algorithm, interpretation and analysis of the results obtained). Fourth Chapter includes : the most important conclusions of the researcher and her recommendations for future studies.
