Istanbul Kültür University DSpace Openaccess Platformu

Search Results

Now showing 1 - 10 of 25

Metadata only
Cross Mark Coordinate Determination and Automatic Registration for Offset Printing
(2018-08) Kasapoğlu, N. Gökhan; Gergin, Zeynep; Gençyılmaz, Mehmet Güneş; Torbalı, Ayşe Bilge; YÜKSEKTEPE, FADİME ÜNEY; GÜNDOĞDU, FATMA KUTLU; 141772; 30141; 108243; 273471
Offset printing is the method for producing commercialized printed media as newspapers and magazines. There are various factors contributing to the overall print quality such as; paper, ink pigment penetration, ink water balance, hygiene of environment, air temperature and humidity. Moreover, quality control is still operator dependent process where the operator is responsible for the visual checks of the printed material. One of the major checks applied is the correct registration of the plates in order to have a sharp image. The misregistration check may require more than one iterations creating setup time variation. Consequently, this highly time consuming setup takes considerable amount of time, especially in low circulation amounts. In this study a novel method called Cross Mark Coordinate Determination and Automatic Registration (CMCDR) is proposed for setup time reduction. CMCDR is based on x and y intensity profiles of registration marks and applied in a printing shop for automatic registration. For this purpose a traditional registration cross for overlapped color components (CMYK) as well as one registration cross for each individual color component is used to determine misclassification error. Using x and y intensity profiles of individual color component to determine misregistration errors and corrections of misregistrations are reduced with only one iteration.
Metadata only
Data Mining Approach for Quality Control Process Improvement
(2019-09) Gergin, Zeynep; Buldanlı, B.Berk; Şahin, Turabi; Elçi, Lütfi; Ekinci, Mert; YÜKSEKTEPE, FADİME ÜNEY; 141772
While striving for serving high quality products, companies are also struggling for cost efficiency. In other words, today’s competitive business environment entails the fact that the economic benefits of the company must be considered together with the product requirements to be met for customer satisfaction. Hence, companies must focus both on improving their ongoing process and establishing cost efficiency. To accomplish this success, they are trying various ways, and data science and data mining tools are the latest solutions that the companies are using in this digital transformation era. With this motivation, this research is conducted in one of the leading bus manufacturers in the automotive industry for improving the cost efficiency of quality control processes with the application of data mining methods. The company has decided to make optimization in the current routes of test drives. Currently, the busses are sent to various routes for validation after the quality control processes. The company supports this project for identifying optimum route assignments in order to minimize test-drives distances, and consequently decrease the related costs. The study starts with the collection of data from the quality control records. Then, it continues with pre-processing and analysis of data to understanding the quality control (QC) process and failure modes. After that, data are processed on WEKA data mining software, for understanding and inspecting the patterns and relationships between route requirements and specific error codes. Finally, the appropriate classification algorithm is selected. After the application of appropriate data mining algorithm, rules are set for route characteristics, and assignments are done between routes and QC error codes. Optimization of routes is done by considering the minimization of distance that is completed on test drives. The comparative results display 18.11% decrease in fuel consumption after the optimization of the routes with the rules set. A decision support app is also developed on android studio, which can be used by the company for faster route assignment decisions.
Embargo
Optimization Based Tumor Classification From Microarray Gene Expression Data
(Public Library Science, 185 Berry St, Ste 1300, San Francisco, Ca 94107 USA, 2011-02-04) Dağlıyan, Onur; Kavaklı, Halil; Türkay, Metin; YÜKSEKTEPE, FADİME ÜNEY; TR108243; TR40319; TR24956
Background: An important use of data obtained from microarray measurements is the classification of tumor types with respect to genes that are either up or down regulated in specific cancer types. A number of algorithms have been proposed to obtain such classifications. These algorithms usually require parameter optimization to obtain accurate results depending on the type of data. Additionally, it is highly critical to find an optimal set of markers among those up or down regulated genes that can be clinically utilized to build assays for the diagnosis or to follow progression of specific cancer types. In this paper, we employ a mixed integer programming based classification algorithm named hyper-box enclosure method (HBE) for the classification of some cancer types with a minimal set of predictor genes. This optimization based method which is a user friendly and efficient classifier may allow the clinicians to diagnose and follow progression of certain cancer types. Methodology/Principal Findings: We apply HBE algorithm to some well known data sets such as leukemia, prostate cancer, diffuse large B-cell lymphoma (DLBCL), small round blue cell tumors (SRBCT) to find some predictor genes that can be utilized for diagnosis and prognosis in a robust manner with a high accuracy. Our approach does not require any modification or parameter optimization for each data set. Additionally, information gain attribute evaluator, relief attribute evaluator and correlation-based feature selection methods are employed for the gene selection. The results are compared with those from other studies and biological roles of selected genes in corresponding cancer type are described. Conclusions/Significance: The performance of our algorithm overall was better than the other algorithms reported in the literature and classifiers found in WEKA data-mining package. Since it does not require a parameter optimization and it performs consistently very high prediction rate on different type of data sets, HBE method is an effective and consistent tool for cancer type prediction with a small number of gene markers.
Metadata only
Synchronized Two-Stage Lot Sizing and Scheduling Problem in Automotive Industry
(Springer-Verlag Berlin, Heidelberger Platz 3, D-14197 Berlin, GermanySpringer-Verlag Berlin, Heidelberger Platz 3, D-14197 Berlin, Germany, 2012) Özdemir, Rifat Gürcan; YÜKSEKTEPE, FADİME ÜNEY; 141173; 108243
This study proposes a mixed-integer linear programming model to solve the synchronized two-stage lot sizing and scheduling problem with setup times and costs. The main motivation behind this study is based on the need of improvement in the inventory control and management system of a supplier that serves to automotive industry. The supplier considered in this study has synchronized two-stage production system which involves three parallel injection machines in the first stage and a single spray dying station in the second stage. In order to achieve synchronization of the production between these stages and improve the inventory levels, a mixed-integer linear programming model is proposed. By the help of developed model, synchronization between the parallel injection machines and the single spray dying station with minimum total cost is obtained. Optimal lot sizes for each stage and optimal schedules of these lot sizes are determined by the developed model as well.
Metadata only
E-Ticaret Müşteri Davranışını Tahmin Etmek İçin Bir Veri Madenciliği Yaklaşımı
(2018-06) Altunan, Büşra; Arslan, Ebru Dilara; Seyis, Merve; YÜKSEKTEPE, FADİME ÜNEY; 108243
1841 yılında kurulan Watsons, 11 farklı pazarda 6300'den fazla mağaza ile dünyanın önde gelen güzellik ve kişisel bakım endüstrilerinden biridir.280 Watsons mağazasına ek olarak, online alışveriş de Türk müşterileri için bir alternatiftir. Mevcut eğilimler nedeniyle, birçok müşteri online alışverişi tercih etmektedir. Müşterilerden bazıları, ürünlerini market sepetlerine eklemekte, ancak maalesef, satın almadan web sitesinden ayrılmaktadırlar. Bu durum, e-ticaret perakendecilerinin çoğu için önemli bir soruna neden olmaktadır. Bu projede, web sitesi ziyaretleri sırasında müşterinin davranışını tahmin etmek için bir veri madenciliği yaklaşımı gerçekleştirilecektir. İlk adımda, web sitesi müşterisinin belirli bir gün için demografik ve davranışsal verileri toplanacaktır. İkinci adımda, veriler ön işlemeden geçirilecek ve eksik değerler kontrol edilecektir. Bir müşterinin satın alıp almayacağını öngörmek için önemli özellikler belirledikten sonra, en doğru veri sınıflandırma yöntemi WEKA kullanılarak belirlenecektir. Web sitesini ziyaret eden müşterilerin alışveriş yapıp yapmama konusundaki eğilimini tahmin etmek için bir karar destek sistemi önerilecektir. Sonuç olarak, şirket müşterinin davranışını web sitesine ilk girişinde tahmin etmesine faydalı olacak bir yöntem önerilecektir.
Metadata only
Investigation of New Facility Location Fır Elvan Gıda
(2019-06) Dinçer, Tutku; Tuncer, Müge; Özgür, Müge; İnce, Buse; Bayramoğlu, Feyzanur; Altınyar, Aydan; YÜKSEKTEPE, FADİME ÜNEY; 108243
Elvan Gıda, established in 1952 as a small candy shop in Istanbul, started to manufacture sweet and chocolate to meet high quality confectionery demand with affordable prices. Elvan has 7 different production facilities located in Istanbul (5), Sakarya and Eskişehir. Currently, company is dealing with a problem of gathering all facilities in a single facility to eliminate fixed operating costs of each while covering desired increased capacity. In this study, an analytical approach will be proposed to solve the facility location problem of the company. First of all, the necessary data about distribution channels of finished goods, transportability, capacity of current and planned facilities, desired optimal distance to Ambarlı harbor and planned budget will be collected. According to determined criteria’s, AHP methodology will be used for selecting 5 promising potential alternatives by considering the executive boards’ evaluations. As a next step, a mathematical model will be developed to decide the least cost location alternative by considering both inbound and outbound transportation costs. Consequently, the optimal facility location with an aggregate layout plan will be proposed to ELVAN company.
Metadata only
A Decision Support Tool for Classification of Turkish SMEs’ Industry 4.0 Score Levels
(Springer Science and Business Media Deutschland GmbH, 2021) Dündar, Uğurcan; YÜKSEKTEPE, FADİME ÜNEY; GERGİN, ZEYNEP; EMİR, OĞUZ; Gençyılmaz, Güneş M.; Çavdarlı, Ali İhsan
The concept of industry 4.0 aims to enhance the employment of digitalization with the help of burst in computer usage after the third major revolution in industry. Hence, to keep up with the new trends in industry, companies should transform their processes into digital platforms. In Turkey, 99.8% of the companies are SMEs, so in order to be in the global economic competition Turkey should enable their SMEs to make the process shifts to digitalization. In this manner, in 2018 a project is initiated by KOSGEB to evaluate the readiness of SMEs and help them for Industry 4.0. As the last step of this project, a decision support tool is created with classification algorithm behind. In this paper, chosen instruments are used to predict the Industry 4.0 score of SMEs that are calculated with previously published algorithm in the scope of the project. In terms of classification, three algorithms are compared with ROC metric. CatBoost algorithm which is specifically created for categorical classification is compared with Support Vector Machine algorithms (SVM and µ-SVM) that performed best in previous research. CatBoost outperformed other algorithms and is used as base classification method in decision support tool. This decision support tool will be used in decision making process of KOSGEB to which companies to help. © 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG.
Metadata only
A Mathematical Programming Approach to Multi-Way Kidney Transplantation
(2015-07) Kutlu, Fatma; Girgin, S.; YÜKSEKTEPE, FADİME ÜNEY; AKTİN, AYŞE TÜLİN; 273471; 108243
Metadata only
A Comparative Sectoral Analysis of Industry 4.0 Readiness Levels of Turkish SMEs
(Springer Science and Business Media Deutschland GmbH, 2021) EMİR, OĞUZ; GERGİN, ZEYNEP; YÜKSEKTEPE, FADİME ÜNEY; Dündar, Uğurcan; Gençyılmaz, Güneş M.; Çavdarlı, Ali Ihsan
The concept of Industry 4.0 brings new standards and has an essential impact on business models, work organizations, and processes. Nowadays, companies are enforced to perform such challenges as meeting personal customer requests, flexibility, responsiveness, customer-oriented solutions, and delivering continuous value to survive in the competitive market. The most innovative companies are resourceful to integrate new digital tools into their business models by following initial trends. For Small and Medium-Sized Enterprises (SMEs), adopting these new industrial challenges, and responding to them quickly is vital when it is considered that they have 99.8% of the enterprise share of Turkey. To do so, an intensive effort is needed to integrate Industry 4.0 applications within the enterprises. It has been suggested in the literature that manufacturing organizations should begin with understanding their current level of maturity by defining their strengths and weaknesses. Further, a significant effort is required with the collaboration of government, academia, and industry leaders to guide enterprises for improving their capabilities in an objective and standardized manner. Despite the diversity of approaches directed to Industry 4.0 concepts, it is not easy to find a study that measures how the manufacturing sector is adopting Industry 4.0 and compares manufacturing sectors with each other. To this end, broad research is initiated in March 2018 to identify the current situation of SMEs in Turkey. In this paper, it is intended to extend the study by providing a comparative sector analysis for manufacturing SMEs in Turkey in terms of Industry 4.0 transformation. Possible action plan suggestions are presented according to the results obtained at the end of the paper. © 2021, The Editor(s) (if applicable) and The Author(s), under exclusive license to Springer Nature Switzerland AG.
Metadata only
Selection of BRCA1/2 negative cases using data mining analytical approach for hereditary breast cancer prediction in high risk breast cancer patients
(Amer Assoc Cancer Research, 615 Chestnut St, 17th Floor, Philadelphia, Pa 19106-4404 USA, 2015-08-01) Kılıç, Seda; Şükrüoğlu, Özge; Akdeniz Ödemiş, Demet; Avşar, Mukaddes; Tunçer, Şeref Buğra; Aysever, Şimal; Güreli, Suzan; Yazıcı, Hülya; YÜKSEKTEPE, FADİME ÜNEY; AKTİN, AYŞE TÜLİN; 140724; 107960; 195520; 108243; 109203; 1292
Background:The causes of hereditary breast cancer are classified in two groups as modifiable and non-modifiable (genetic) in our study.Some of the non-modifiable factors selected are breast density, menarche age, menopause, BRCA1 and BRCA2 genes, family history, height and weight of patients.On the other hand, oral contraceptive usage, alcohol consumption, hormone therapy, breastfeeding, exposure to radiation and smoking are among the modifiable factors.The risk of having hereditary breast cancer could be identified by BRCA1/BRCA2 gene tests which requires long lasting experiments and incurs high costs. Before performing the required genetic test, the modifiable and non-modifiable factors of the high risk patients are collected by geneticist or genetic counselor and then calculated risk score of the patients.After observing the test results, the effect of existing factors are analyzed.The aim of this study is to develop a hereditary breast cancer prediction algorithm by using data mining techniques for high risk breast cancer patients.Methods:Different applications may require different data mining methods.The current study involves the data classification technique for predicting hereditary breast cancer through a function that incorporates modifiable and non-modifiable factors. The data consists of 562 BRCA negative, and 68 BRCA positive instances with 117 categorical and numerical characteristics.The first step of this approach is preprocessing the data on hand.Individuals with incomplete data are excluded from the list, and some logical adjustments are performed.Results:As a result, 440 patients (392 BRCA negative, 48 BRCA positive), and 75 characteristics (3 numerical, 72 categorical attributes) are obtained.In the second step, important factors are determined by attribute selection algorithms. After applying 23 selection and 7 ranking methods, the important attributes are determined as: age, diagnosis, height, weight, menopause age, menarche age, FIB/FIB<40, SEB/SEB<40, THB/THB<40, FIO/FIO>40, SEO, cancer status of the family, clinical stage, and pathological stage.As a result, three data sets are suggested for classification. Finally, a total of 182 data classification results are derived for the aforementioned data sets.All analyses are performed by employing WEKA software.Conclusion:Comparison of the results shows that, the best data classification method's predictions on the BRCA negative class has 100% accuracy.Hence, a new individual's situation can be assessed without undergoing detailed gene tests, resulting in cost and required workforce reduction.The results are found to be promising and applicable.As an extension of the study, a user friendly interface and a decision support tool can be developed.

Filters

Settings

Sort By

Results per page

Search Results