14
Archived Theses
0
DOIs Assigned
0%
DOI Rate
Discipline
Hava yolu gelir yönetiminde yolcuların uçuşa gelme oranının tahmini için bir bulanık yaklaşım
Revenue Management logic had been firstly used in aviation and then hotel sectors has still had big place in airlines' strategies increasing their revenue. Airline companies enabling market segmentation via Pricing and Yield Management methodologies aim to enlarge the number of passengers reached and obtain the maximized revenue. Overselling which is one of the foremost points of Yield Management is a very important income tool for airlines whose product is perishable and meaningless after departure. The possible cost revealed because of no-show passengers is compensated by overbooked customers. Show-up rates of the passengers are investigated in this thesis via six factors; fare, number of direct flights between origin and destination points, difficulty of having visa from destination country, human development index of destination country, distance and temperature and a forecast model is offered. There are six-month data instead of one whole year and the data is comprised of eight destination points, fuzzy linear regression method which is better in uncertain conditions and human perception is opted rather than statistical linear regression. At the end of application, show-up rate of a passenger is calculated between a minimum and maximum value for each line with respect to its characteristic. The main aim of the forecasting model is to predict the show-up rate of the passengers and minimize the error of number of overbooked customers for a brand-new airline.
Üçüncü parti lojistik hizmet sağlayıcı seçimi için bir bulanık çok ölçütlü karar verme yaklaşımı
Nowadays, supply chain management becomes more important day by day as a result of rapid competition and need for cost reduction in supply chain operations that include logistics activities. Supply chain departments of firms have become major elements instead of support elements in the global economy. This situation also leads to more risk-oriented supply chain operations and Supply Chain Risk Management notion is aroused from focusing not only benefits of the suppliers but also risks and uncertainties in the long term. 3PL Provider Selection is classified as a supplier selection problem is an important part of supply chain management. It is important to evaluate both positive and negative sides of suppliers in the supplier selection process. The aim of this study is arranging in order the criteria that are used in 3PL provider selection problems via DEMATEL Method and enhancing a model that assesses the risk of alternatives via applying Fuzzy TOPSIS Method with the help of the criteria that specifies risk. The relationship between the assessment of alternatives according to criteria that is made by experts and risk factors that alternatives have is fuzzy. Fuzzy TOPSIS Method enables to choose the most appropriate 3PL firm in the long run. A real case from cosmetic company is applied to demonstrate the efficiency of the proposed methodology.
Risk faktörleri gözönünde bulundurularak tedarik zinciri yönetimi: Vaka çalışması
Abstract : Today's competitive business world make supply chain management (SCM) much more complex and dynamic. Besides this competitive environment, growing uncertainties also impact firms seriously and force them to develop sustainable approaches along supply chain (SC) operations. The aim of this study is constructing a comprehensive configuration to help decision makers during the supply chain management process and simplify to understand relationships among Supply Chain Integration (SCI), Supply Chain Strategies (SCSs), Supply Chain Risk Factors (SCRFs) and Performance Criterias (PCs). Criteria of SCI, SCSs and PCs are based on literature rewiev. SCRFs is also based on literature but our reference study about SCRFs includes 20 RFs and they are reduced to 4 main RFs by experts to compose more compact model. Our experts are working in automotive sector in Turkey. First, Supply Chain Configuration (SCC) is designated in this survey. Then, direction and strength of relationship among concepts are identified according to experts' answers and all relationships are examined by the means of Fuzzy Logic and Fuzzy Cognitive Map (FCM) methodologies. The results, paralelled to literature, show that proposed SCC can help us to analyze relationships from the broad perspective and it also enables to react risks immediately. Keywords : supply chain management, risk factors, supply chain integration, supply chain strategies, performance criterias, FCM, fuzzy logic
Dental implant imalat endüstrisinde ürün iyileştirmesi için kalite fonksiyon yayılımı temelli karar verme yaklaşımı
Quality function deployment (QFD) is a great tool in translating customer needs into specifications into each phase. Based in Japan, QFD quickly spread to U.S. where it gained most of the popularity. This study subjects a SME in Turkey, who has ongoing research about the most important aspects to supply in dental implant industry before entering the market. Since the resources are limited, a study is going on to find the most important processes complying with customer needs and transferring them to manufacturing operations. Study includes a two-step QFD approach to for translating customer needs into quality characteristics. First step starts with extensive clinical and marketing research to figure out customer needs. This constructed the first WHATs of QFD chart. Second step contains using QFD in dental implant manufacturing. This step starts with another meeting between engineers and customers to find out design characteristics. This time customer needs turn into WHATs and design characteristics became HOWs. This step continues with transferring such aspects into manufacturing most important manufacturing operation is found. Finally, the relations between quality controls and manufacturing operations were inspected and the most important element is determined. Results are R&D, short length, CNC machining and geometric inspections in the most important process, quality characteristic, manufacturing process and quality controls respectively.
Karar verme yöntemlerinde bulanık sistem genişlemeleri ve uygulaması
Cognitively speaking, some processes or steps included in decision- making can be a serious burden for DMs. Pairwise comparisons are one of those steps to be handled carefully in order to obtain useful results. During pairwise comparisons of elements such as decision criteria, let alone providing exact quantitative judgments, a complete linguistic set of judgments may sometimes be hard to retrieve from DMs. From that perspective, completing an incomplete pairwise comparison matrix is an interesting problem that many attempted to cover. For that purpose, when DMs provided incomplete interval-valued fuzzy preference relations, experimental approach was used to obtain the completed matrix. Also, the fuzzy synthetic extent was used to calculate the fuzzy weights of criteria. The purpose of this exploratory study is to analyse what happens when triangular fuzzy numbers (TFNs) are used for judgments and how powerful is the method in order to represent the will of the DMs. In the study, a numerical application is provided using incomplete pairwise comparison matrices of dimension four and five collected from 40 decision makers where their preferences are given using a linguistic set. The Accuracy of the approach is then tested for different levels of confidence in order to provide insightful concluding remarks.
A new mathematical programming formulation for multivariate regression clustering with a store clustering application in retail sector
Clustering in machine learning is an unsupervised learning technique, and it is the study of grouping similar entities together. In particular, clustering via regression groups entities such that the fit of a linear-in-parameter model is "best" with respect to a norm function of residuals---usually an L^2-norm (Euclidean norm) of residuals, which gives the least-squares estimates of the unknown regression parameters. This study considers linear-in-parameters regression model-based clustering when there are multiple correlated responses (i.e., dependent variables). Furthermore, this study proposes two new mixed-integer linear programming formulations, where the first formulation is a simple extension of the classic formulations with the big-M constraints in the literature for multiple responses. Besides, the objective function minimizes L^∞-norm of residuals instead of L^1-norm or L^2-norm. However, the big-M in the formulations depends on the unknown clusters and regression parameters, and hence, it is impossible to know a "good" value for the big-M a priori. A too big value for the big-M results in numerical instabilities, and a too small value for the big-M cuts the true optimal solution. Therefore, this study proposes to replace the classic formulations with the big-M constraints with formulations that have special ordered sets of type one (SOS Type-1) variables. The proposed formulation for multiple responses with the big-M constraints is applied to the JURA dataset, which clusters locations to predict the concentration of some metals that are more expensive to measure using measurements of metals that are cheaper to sample. For this numerical example, this study illustrates that the classic approach in the literature, which considers one response at-a-time, usually assigns the same entity to different clusters with respect to different responses; hence, eventually, it is not evident to which cluster that entity belongs to. Also, even though the classic approach assigns that entity to the same cluster with respect to all responses, this assignment can be different than the one obtained when all responses are considered simultaneously; hence, the classic approach can result in a false clustering when multiple corelated responses have to be considered at the same time. The proposed formulations and algorithms are also applied for store clustering problem of a fashion retailer in Turkey. The store clustering problem simply means grouping of the stores based on some predetermined characteristics so that those stores within the same group are more similar to each other than the other stores. The purpose of such a clustering can be to detect and analyze the effects of location, customer and store characteristics on the store performance, which can depend on different clusters. The proposed models are solved using the branch-and-cut algorithm using optimization software Gurobi 8.1.0 C API.
Performans değerlendirme için karar verme yaklaşımları
Companies are paying more attention to performance evaluation as markets are dynamic and competition is increasing. Performance evaluation gain importance issue in order to overcome competition and identify their strategies. This approach with improving performance evaluation criteria in companies, ensured that companies are managed correctly, specify effective strategies for their companies and obtain the advantage of competition. In literature, there are many studies that contain performance evaluation topic but generally, performance evaluation is specified with just financial performance evaluation. In this study, beside the financial performance evaluation metrics, strategic criteria, customer criteria etc. are considered. Balance Scorecard (BSC) is one of the most important methods for performance evaluation and strategic road map. In this research, performance measures categorized with BSC perspective, therefore evaluation metrics include financial and non-financial criteria. This study is specialized with performance evaluation in four dimensions. General performance evaluation model structure is created with Benefits, Opportunities, Costs, Risks (BOCR) method. Criteria weights are calculated with Fuzzy Analytic Network Process. This study is applied in retail sector and alternatives are ranked with Technique for Order Preference by Similarity to Ideal Solution (TOPSIS).
Otomatik insan ruh sağlığı asistanı: Pasif sensör verisinden stres tanıma çalışması
Stress level among people is rising through years and passive sensing data from mobile phones or other ubiquitous devices have started to found its place in applications of mental health observation. With the ultimate goal of creating an automatic human mental health assistant that helps people to have a better mental condition, a step is taken by creating a stress recognition model. In previous works, the researchers have found correlations between sensor data and mental health conditions and attempted to predict the stress level of the user. Due to there is no direct link between any sensor data with mental health, Machine Learning algorithms are employed to uncover relations with multiple sensors and mental well-being. The utilized machine learning algorithms for prediction work with non-sequence data hence the researchers need to extract features that represent historical sensor data with instant features. However, extracted features cannot completely represent a sequence of time data. Within the scope of this study, we showed that LSTM, CNN and CNN-LSTM algorithms which accept sequences of data as input and reaches exceptional performances in different applications can also work in passive mobile phone sensor data to predict human mental stress. The performance of the model on StudentLife dataset which includes passive mobile sensing data of college students has 62.83% accuracy on 460 test instances by training with 800 instances with LSTM model. Diversity and size of the data are very small and the data-hungry LSTM model could not generalize on adapted features with the small sample size. Although we did not adapt complex features, the results are promising and encourage us to improve data size and continue to research on this topic.
Havayolları için yeni rota seçiminde çok kriterli karar verme yöntemi yaklaşımları
In this study, it is aimed to select the new best airport destination for an airline in Turkey with the base situated in Istanbul Airport. This study is unique in terms of plentiful criteria used and a fresh perspective brought by the comparative analysis of BOCR and simple ANP methods. It also fills a gap in the literature caused by the lack of ANP applications for destination selection. In this work, first, a brief review of literature is presented including studies using multi-criteria decision-making in route evaluation and in the airline industry. The model is constructed both using one-level Analytic Network Process (ANP) and Benefit, Opportunity, Cost and Risk (BOCR) analysis. Both normalized statistical direct data and expert opinions based on the 1-9 scale of Saaty are used to determine the weight of each criterion through analysis. According to simple ANP method, Mariscal Sucre International Airport in Tababela, Quito is the best alternative with the strategic considerations as the most dominant criterion, while BOCR analysis shows that Denpasar, I Gusti Ngurah Rai International Airport is the best under benefit, opportunity and risk dimensions with the passenger demand as the most dominant criterion and as for the cost perspective, Prince Said Ibrahim International Airport situated in Moroni is the best, while it is the worst alternative under risk dimension
VZA kullanarak uçuş bazında ilişkisel karbon emisyon performans analizi: Bir vaka çalışması
Governments and international agencies act to prevent pollution. One these actions is limitation on greenhouse gases emission to prevent global warming. Today, world is fully industrialized and industries requires energy to operate. Energy creation from fossil fuel combustion, releases carbon dioxide into the atmosphere. Carbon dioxide is the major perpetrator of global warming. An aviation company should embrace the required carbon emission limitation acts if it wants to operate profitable as it is now. Recent studies and actions of government and authorities indicates that there will be high fines to polluters because of excessive amount of emission. One of examples for these applications is EU Emission Trading System (ETS) is already in action since 2012 for aviation industry operates in Europe and charging for excessive amount of carbon emission. The aim of our study is to provide a model to detecting improvement potential on flights by assessing carbon emission performance relatively for flights in a period. At the beginning, we ask 12 aviation experts 'which factors affect the emission amount produced by a flight?', and ask them to rank these factors using a fuzzy linguistic matrix we provide. Using historic flight data and weights, which are calculated using the insights from our experts, we construct and apply two DEA based on constant return to scale (CRS) and variable return to scale (VRS) on our empirical dataset. The results, similar with literature, showed our models would help environmental improvement responsible to determine improvement potentials.
Türkiye çiçekçilik endüstrisi için sürdürülebilir kalkınma stratejilerinin ve sürdürülebilir finansman mekanizmalarının değerlendirilmesi
Sustainability concept, which is mostly associated with environmental approaches, has evolved into policies under the title of sustainable development throughout the world, and has been transformed into an action plan that will cover the present and future generations by addressing the economic and social dimensions. It is clear that agriculture plays an important role in sustainability. Floriculture, which is one of the significant agriculture-based industries, maintain its importance for many countries with its commercial position. Although Turkey has significant advantages, it can be seen that it's not a major player in the global floriculture market. In addition, there are serious public debates for quite some time, about the decline in the agriculture. The aim of this study is to determine strategies for the sustainable development of the Turkish floriculture industry, and thereupon to evaluate these strategies and sustainable financing mechanisms together with their economic, environmental and socio-political dimensions. Within the scope of the study, we have identified sustainable development strategies that can be applied to the industry, with the help of a comprehensive Strength-Weakness-Opportunities-Threats analysis, where we highlight the current situation of the Turkish floriculture industry, and individual expert opinions. The Analytical Network Process model that we developed in line with our goal, makes it possible to evaluate the sustainable floriculture approach under benefits, opportunities, costs and risks merits. We believe that the criterion system that we have identified and results we obtained, will draw an applicable strategic road map on behalf of Turkish floriculture and other similar industries.
Otomotiv üreticileri için esneklik portföyü
Nowadays, manufacturing companies are in serious competition and in an uncertain environment. As business conditions fluctuate and technological developments occur, customers are less predictable in purchasing behavior. In order to survive in the competition and to increase the market share, flexibility is proposed in this study. The automotive sector is one of the sectors that flexibility is highly needed. The purpose of this study is to find out the best flexibility lever portfolio satisfying customer needs considering also supplier and retailer expectations and providing a quick solution to deal with the uncertainties and the risks this company faces for automotive manufacturer. As a methodology, a decision making model is developed by combining analytic network process (ANP) and quality function deployment (QFD). Firstly, supplier, retailer and customer expectations are ranked within themselves, and then according to automotive manufacturer their importances are ranked via ANP, secondly the flexibility levers matching the expectations are weighted using two-phased QFD that is house of flexibility concept. In here, a house of flexibility is constructed to relate supplier, retailer and customer expectations with flexibility levers as a first phase. In this part of study, the best portfolio that all expectations can be satisfied maximum with constraint of minimum flexibility is proposed. Later on, a second house is built to detail down the flexibility levers into system factors. As a result, quick solutions are provided to deal with uncertainties and the risks faced by company.
Bulanık nötrosofik EDAS yöntemi ile Körfez ülkelerindeki 3. parti soğuk zincir tedarikçileri için depo yeri seçimi
From past to present, logistics processes have been analyzed as one of the most important expense items for companies. And the companies that failed to manage their logistics processes well, cannot reach the desired growth rates and some even disappeared. Logistics management without process optimizations can be time and money consuming for the companies. So as it is clear, logistics processes have a significant influence on organizations' efficiency. That's why the process flow quality should be kept at a certain level and some decision problems should be analyzed and answered well. Location selection of a warehouse is one of the most important decision problems of supply chain and logistics management. Since different alternatives need to be analyzed, MCDM techniques are used to solve this problem. And alternatives are evaluated in terms of both quantitative and qualitative criteria to decide the best location alternative by using scientific MCDM techniques. Over the years different kinds of scientific methods have been developed and used to solve this problem. As for this study, the aim is to investigate a more specific topic of warehouse location selection. This specific topic is the warehouse location selection of third-party cold chain logistics suppliers. Recently, the importance of keeping perishable products under right conditions increased, since it is realized that cold chain warehousing and transportation prevent the product/food spoiled. So organizations can save their money and time by having an effective cold chain management. The main scope of this study is to give a new scientific point of view for warehouse selection in cold chain logistics. MCDM technique of Interval Valued Neutrosophic EDAS method has been used to evaluate important criteria and to choose the best alternative. And some important cities of Gulf Region countries have been handled and evaluated as alternatives in the computational study of this project. Du passé au présent, les processus logistiques ont été analysés comme l'un des postes de dépenses les plus importants pour les entreprises. Et les entreprises qui n'ont pas bien géré leurs processus logistiques ne peuvent pas atteindre les taux de croissance souhaités et certaines ont même disparu car ces processus peuvent prendre du temps et de l'argent si les optimisations de processus n'ont pas été faites. Il est donc clair que les processus logistiques ont une influence significative sur l'efficacité des organisations. Étant donné que différentes alternatives doivent être analysées, les techniques MCDM sont utilisées pour résoudre ce problème. Et les alternatives sont évaluées en termes de critères quantitatifs et qualitatifs pour décider de la meilleure alternative de localisation en utilisant des techniques scientifiques MCDM. Au fil des ans, différents types de méthodes scientifiques ont été développés et utilisés pour résoudre ce problème. Quant à cette étude, elle vise à étudier la sélection de l'emplacement des entrepôts pour la logistique de la chaîne du froid. Récemment, l'importance de conserver les produits sensibles à la température dans de bonnes conditions a augmenté, car l'entreposage et le transport dans la chaîne du froid empêchent le produit / l'aliment de se gâter. Ainsi, les organisations peuvent économiser leur temps et leur argent en ayant une gestion efficace de la chaîne du froid. L'objectif principal de cette étude est de donner un nouveau point de vue scientifique pour la sélection des entrepôts dans la logistique de la chaîne du froid. La méthode EDAS neutrosophique à valeurs d'intervalle a été utilisée pour évaluer des critères importants. Et la région du Golfe a été traitée dans l'étude informatique de ce projet.
Özgün bir kapsamlı karar verme yaklaşımıyla gemi inşa sektöründe proje tipi seçimi
The ability of a nation to develop and produce is becoming more and more essential in today's world. The wealth, fortune, and safety of society are directly impacted by this aptitude. Nations must consider their particular peculiarities, such as geographic, environmental, scientific and technical factors, to continue to enhance this capability. The shipbuilding industry deals with the creation of bigger (mostly oceangoing) vessels intended for the merchant fleet (cargo or passenger carriage), military resolutions or pleasure and private life. It also includes goods and services provided for the construction, modification, and upkeep of these vessels. In recent years, it has become clear that one of the industries in which Türkiye excels in the area of meticulous design and production is shipbuilding, which increases our country's exports. In order to ensure greater sustainability and development in this area, it would be valuable to perform a study that will help specify the appropriate initiatives. The main purpose of this study is to ascertain the most felicitous project type by taking into account the interaction between the criteria to appraise the project type selection in a multi-dimensional perspective with the details received from shipyards. In this study, after the criteria are identified, a well-known decision-making model will be used and implemented. High-tech production caliber shipyards can utilize the proposed strategy to decide on a project if adequate information has been gathered.