Fen Bilimleri Enstitüsü

Permanent URI for this communityhttps://hdl.handle.net/20.500.12573/192

Browse

Now showing 1 - 8 of 8

Türkiye'de Elektrikli Araç Sarj İstasyonu Lokasyonu Belirleme
(Abdullah Gül Üniversitesi, Fen Bilimleri Enstitüsü, 2023) Gülbahar, İbrahim Tümay; Sütçü, Muhammed
Electric vehicles are now regarded as one of the best and greenest replacements for internal combustion engine vehicles. For the widespread use of electric vehicles, the construction of the vehicle charging network and, in particular, the choice of the appropriate site for the charging stations, are viewed as critical issues. The majority of studies on the topic concentrate on well-known locations like city centers, shopping malls, and airports. Because there are so many alternative charging stations, even though these and comparable locations are regularly used in everyday life, they can usually provide an appropriate solution to the daily charging need. For intercity travel, it is impossible to find enough charging stations, especially on highways. To choose the position of electric vehicle charging stations on highways, a decision model has been suggested in this study. The anticipated number of electric vehicles in Türkiye over the next few years is projected in order to acquire a realistic approach to the location of charging stations, and this amount is employed as a significant input in the facility positioning model. The best places for charging stations on state highways that can meet customer demand were then identified using an optimization technique. The suggested model selects the most suitable locations for charging stations and the number of chargers that should be installed there while also making sure that drivers of electric vehicles on highways don't run into charging issues.
Sembolik Toplam Yaklaşım Kümelemesi Yoluyla BIST100 Yatırımlarında Yön Bulma: Yatırımcılara Yönelik Bilgiler
(Abdullah Gül Üniversitesi, Fen Bilimleri Enstitüsü, 2024) Nalici, Mehmet Eren; Ünlü, Ramazan; Söylemez, İsmet
Market stakeholders, including traders and investors, strive to forecast stock market returns for informed decision-making. Computational finance employs various tools such as machine learning techniques to analyse extensive financial datasets to provide predictive insights for investors. Among all those techniques, clustering is one of the most well-known and used machine learning methods to reveal hidden patterns from unlabelled data. This study aims to help investors make more robust decisions by autonomously identifying companies that may exhibit similar price movements. In our study, with the model developed based on the Symbolic Aggregate Approximation (SAX) method, BIST100 companies are divided into clusters of various numbers and various scenarios are developed for investors from different perspectives such as risk minimization and strategic investment. The SAX clustering method is employed for analysing share movements. Moreover, dendrogram tree graph is used to analyse the clustering of different SAX combinations.
Fiyat Koruması, Dönem Ortası ve Dönem Sonu Geri Ödemesi Altında Ücretlendirme Stratejileri
(Abdullah Gül Üniversitesi, Fen Bilimleri Enstitüsü, 2022) Yıldız, Barış; Sütçü, Muhammed
In this thesis, we examine a selling environment where a manufacturer-controlled retailer and an independent retailer sell a slow-moving A item. The manufacturer offers the independent retailer price protection against reductions in the wholesale price. The price set by the independent retailer is assumed to be determined by Retail Fixed Markdown (RFM) policy. The manufacturer adopts a periodic-review pricing strategy and each retailer observes price-dependent stochastic demand. We employ Multinomial Logit (MNL) models to forecast customers' preferences based on retail prices. We construct stochastic programming models to determine the manufacturer's pricing strategy in the presence of four distinct price commitment contracts which differ in the supplementary privileges combined with price protection. We also propose a variant Stochastic Dual Dynamic Programming (SDDP) algorithm to determine the manufacturer's approximately optimal pricing strategy by getting around three curses of dimensionality. We observe the impact of critically important contractual parameters on the price, the market shares and the expected true profits. We also evaluate the performance of the proposed algorithm and compare the price commitment contracts in terms of the contractual parameters for which it is crucial to choose a compromise value to ensure high enough profitability for both retailers.
Optimal Isı Yalıtımı için Yapı Malzemelerinin Özelliklerinin Belirlenmesi ve Yapı Malzemesi Seçimi
(Abdullah Gül Üniversitesi, Fen Bilimleri Enstitüsü, 2023) Kılıçarslan, Mustafa Özgür; Kara, Gökmen
The escalating urgency of climate change demands innovative approaches to energy conservation, particularly in the realm of building construction, known for substantial energy consumption and greenhouse gas emissions. This research delves into transformative strategies for enhancing energy efficiency in office buildings, with a concentrated analysis of the implementation of advanced building materials and state-of-the-art construction methodologies. Utilizing OpenStudio, a cutting-edge energy modeling software tool from the U.S. Department of Energy's National Renewable Energy Laboratory, this study quantitatively evaluates the energy-conserving potential of various avant-garde materials and construction techniques. The investigation is anchored around a case study of an office building in Ankara, Turkey, serving as a representative model for exploring diverse scenarios. These scenarios encompass the integration of high-performance framing, airtight construction, materials with superior thermal resistance properties, and advanced glazing systems. The research meticulously assesses each scenario with the aim of delineating the configurations that most significantly reduce energy consumption. The results reveal that specific combinations of advanced techniques and materials can lead to substantial reductions in energy use, thereby contributing profoundly to global efforts in mitigating climate change impacts. The conclusion emphasizes the necessity for widespread adoption and standardization of these energy-efficient practices in the construction industry, proposing them as pivotal contributors to the broader environmental sustainability movement.
Dengesiz Sınıflandırma Sorunlarına Torbalama ve Arttırma Esaslı Yeni Bir Yaklaşım
(Abdullah Gül Üniversitesi, Fen Bilimleri Enstitüsü, 2022) Pınar, Muhammed Şafak; Akgün, İbrahim
Classification algorithms are employed in a wide range of real-world problems such as obstacle detection, fraud detection, medical diagnosis, spam detection, speech recognition, image processing, intrusion detection, and so forth. However, it is not always an easy task to propose a legitimate classifier. For a classification task, there are numerous limitations of datasets. One of the most confronted limitations in real-world classification tasks is skewed class distribution, also called the class imbalance problem. When learning is employed in class imbalanced datasets without incorporating appropriate adjustments into the existing algorithms, minority classes are mostly misclassified. This study introduces a novel classification algorithm that outperforms previous studies on benchmark datasets used for the class imbalance problem. The presented novel algorithm, namely, BagBoost, involves aggregating modified bagging and modified boosting algorithms to increase the visibility of minority class instances. The state-of-the-art algorithms in the classification of imbalanced datasets are investigated. The results of the best existing algorithms are compared with the proposed algorithm using benchmark datasets. Results show that BagBoost is a better classifier than commonly used classification algorithms in the literature for benchmark datasets according to F-measure and G-mean scores.
Ağaç Yapılı ve Ayrıt Kapasiteli Hub Yerleşim Problemleri için Model ve Çözüm Metodolojilerinin Geliştirilmesi
(Abdullah Gül Üniversitesi Fen Bilimleri Enstitüsü, 2022) Kayışoğlu, Betül; Kayışoğlu, Betül; Akgün, İbrahim
In this dissertation, we study two different extensions to hub location problems, namely, Multiple Allocation Tree of Hubs Location Problem (MATHLP) that result from incorporating a tree topology requirement for the hub network and Multiple Allocation Arc Capacitated Hub Location Problem (MACHLP) that result from imposing capacities on the arcs. We consider both problems in a multiple allocation framework and try to minimize total flow cost by locating p hubs. Unlike most studies in the literature that use complete networks with costs satisfying the triangle inequality to formulate the problems, we define the problems on non-complete networks and develop a modeling approach that does not require any specific cost and network structure. Our proposed approach provides more flexibility in modeling several characteristics of real-life hub networks. We solve the proposed models using CPLEX-based algorithm and Gurobi-based algorithm with NoRel heuristic. For MATHLP, we develop Benders decomposition-based heuristic algorithms and for MACHLP, we develop a heuristic algorithm based on simulated annealing. We conduct computational experiments using problem instances defined on non-complete networks with up to 500 and 400 nodes for MATHLP and MACHLP respectively. The results indicate that the proposed solution methodologies are especially effective in finding good feasible solutions for large instances. Keywords: hub location problem, tree of hubs location problem, arc capacitated hub location problem, benders-type heuristics, simulated annealing
Belirsiz İşlem Süresine Tabi Paralel Makine Çizelgelemeleri
(Abdullah Gül Üniversitesi, Fen Bilimleri Enstitüsü, 2017) Bekli, Rahime Şeyma; Gören, Selçuk
Günümüz dünyasında iş ve üretim rekabeti, firmaların verimlilik artıran ve imalat maliyetini düşüren çizelgeler üretmesine yol açmıştır. Ancak, üretilen çizelgeler beklenmedik aksaklıklar yüzünden, genellikle amaçlandığı şekilde uygulanamamaktadır. Bu aksaklıklar makine arızalanması, sipariş iptali gibi örneklendirilebilir. Aksaklıklara duyarsız çizelge olan gürbüz çizelgeleme, son yıllarda araştırmacılar arasında önem kazanmıştır. Bu çalışmada, belirsiz işlem süresine tabi paralel makine ortamı ele alınmıştır. Performans ölçütü son işin bitiş süresi olarak alınmıştır. Belirsizlik, ayrık senaryolar olarak modellenmiş ve küçük boyuttaki problemleri çözebilen bir tam sayılı programlama oluşturulmuştur. Bu model büyük problemleri çözmede sıkıntılıdır. Bu sebeple senaryo sayısını azaltma yaklaşımı denenmiştir. Daha sonra eşiz ayrıştırma yöntemi ile büyük problemlerin çözümü amaçlanmıştır. Bu yöntemi kullanmadaki amaç büyük bir problem çözmek yerine, küçük ama çok sayıda problem çözerek sonuca ulaşmaktır. Ancak bu yöntem de büyük problemlerde istenilen sonuçları vermemiştir. Bu sebeple senaryo sayısı azaltılarak eşiz ayrıştırma yöntemi kullanılmış ve yeni bir sezgisel önerilmiştir. Aynı zamanda bir tabu arama algoritması oluşturulmuştur. Sonuçlar, önerilen sezgisel algoritmalardan senaryo azaltılması ve tabu arama algoritmalarının paralel makine ortamında iyi sonuçlar verdiğini göstermektedir.
Türkiye Süper Futbol Ligi'nin Çizelgelenmesi
(Abdullah Gül Üniversitesi, 2019) BAYRAK, HASAN; Bayrak, Hasan; Sütçü, Muhammed
Günümüzde spor etkinliklerinin ekonomik büyüklükleri nedeniyle bir işletme olarak organize edilmesi, pazarlanması ve yönetilmesi gerekiyor. Takımlar yeni oyuncuları transfer etmek için inanılmaz yatırımlar yapıyor. Populer liglerin yayın hakları yüz milyonlarca dolara satılabiliyor. İyi bir turnuva çizelgesi; adalet, organizasyonun güvenliği, maçların seyirci çekmesi, yayıncıların ve reklamcıların yatırımlarının karşılığını alabilmesi açısından oldukça önemlidir. Ancak, adil ve uygun bir program oluşturma niyeti, paydaşların farklı ihtiyaçları ve bu ihtiyaçların sürekli birbiriyle çatışıyor olması nedeniyle kolay değildir. Turnuvadaki takım sayısı arttığında, geleneksel yöntemlerle en uygun çözüme ulaşmak mümkün olmayabilir. Bu tezde, merdiven deseni dediğimiz bir dizi desen eleme yöntemi bulduk. Dahası, merdiven deseni metodu yardımıyla, çok karmaşık bir problemi kolay ve hızlı bir şekilde çözebilecek 5 adımlı bir yaklaşım geliştirdik. Sonuç olarak Türkiye Süper Futbol Ligi için daha iyi bir çizelge oluşturduk.

Browse

Browsing Fen Bilimleri Enstitüsü by Department "AGÜ, Fen Bilimleri Enstitüsü, Endüstri Mühendisliği Ana Bilim Dalı"