Berk Ayvaz

A hybrid type-2 fuzzy based supplier performance evaluation methodology

Display OmittedThe main steps of the developed hybrid methodology. A hybrid fuzzy MCDM methodology is designed to deal with uncertainty resulting from subjective judgments.We present a holistic approach for supplier performance evaluation problem in aviation maintenance industry.The proposed methodology is applied on a leading global aviation company to prove its merit in real life context. In the globalized world, fatal competition ruling in almost all business sectors hits especially airline companies due to the global nature of the industry. Thus, maintaining the business operations in a cost-effective manner is of paramount importance in this destructive business environment. While realizing this goal, a special focus on Maintenance Repair and Overhaul (MRO) operations is indispensable, considering that a significant proportion of expenses in the industry are originated from them. Since the MRO departments are responsible for maintenance, repair, and overhaul of aircrafts and engines under very strict standards, a smooth, clean-cut supply chain does not only provide a solid technical ground for daily operations but also it helps to gain a comparative advantage in cost leadership which, in turn, results with high customer satisfaction. Obviously, a key component of this chain is designing a robust and reliable Supplier Performance Evaluation (SPE) methodology where numerous tangible and intangible evaluation criteria come into play reflecting the multifaceted characteristic of the decision problem. This work proposes a three-phase hybrid approach to address the SPE problem in aviation industry where the suppliers are obliged to follow tight tolerances regarding quality and delivery times. Starting with selection of the most significant evaluation criteria, in the second stage, an Interval Type-2 Fuzzy (IT2F)-Analytic Hierarchy Process (AHP) is employed to determine their relative importance. These weights are then given as input to an IT2F-Technique for Order of Preference by Similarity to Ideal Solution (TOPSIS) method. We used the IT2F extensions due to the fact that some of the evaluation criteria are subjective and qualitative in nature. To prove the merit of the proposed methodology, an application is presented to SPE problem at Turkish Technic Inc., a subsidiary of Turkish Airlines.

Electricity consumption forecasting for Turkey with nonhomogeneous discrete grey model

ABSTRACT The accuracy of forecasting is an essential issue for decision makers in terms of energy planning. During the recent years, several techniques have been used for electricity consumption forecasting in order to accurately predict the future demand. Although there are several forecasting techniques, selection of the most appropriate one is of paramount importance. In this study, three different grey forecasting models are built and used for modeling and predicting yearly net electricity consumption in Turkey. Additionally, these three models are compared to find the best model by using performance criteria. The best approach, Nonhomogeneous Discrete Grey Model (NDGM), is employed to forecast electricity consumption from 2014 to 2030. In addition, a comparison is made with recent studies proving the grey model (GM) proposed by this study delivers better forecasting performance.

Energy-related CO2 emission forecast for Turkey and Europe and Eurasia: A discrete grey model approach

Purpose The global warming, caused by the anthropogenic greenhouse gases, has been one of the major worldwide issues over the last decades. Among them, carbon dioxide (CO2) is the most important one and is responsible for more than the two-third of the greenhouse effect. Currently, greenhouse gas emissions and CO2 emissions – the root cause of the global warming – in particular are being examined closely in the fields of science and they also have been put on the agenda of the political leaders. The purpose of this paper is to predict the energy-related CO2 emissions through using different discrete grey models (DGMs) in Turkey and total Europe and Eurasia region. Design/methodology/approach The proposed DGMs will be applied to predict CO2 emissions in Turkey and total Europe and Eurasia region from 2015 to 2030 using data set between 1965 and 2014. In the first stage of the study, DGMs without rolling mechanism (RM) will be used. In the second stage, DGMs with RM are constructed where the length of the rolling horizons of the respected models is optimised. Findings In the first stage, estimated values show that non-homogeneous DGM is the best method to predict Turkey’s energy-related CO2 emissions whereas DGM is the best method to predict the energy-related CO2 emissions for total Europe and Eurasia region. According to the results in the second stage, NDGM with RM (k=26) is the best method for Turkey while optimised DGM with RM (k=4) delivers most reliable estimates for total Europe and Eurasia region. Originality/value This study illustrates the effect of different DGM approaches on the estimation performance for the Turkish energy-related CO2 emission data.

Towards an autonomous human chromosome classification system using Competitive Support Vector Machines Teams (CSVMT)

Abstract In broad terms, karyotyping is the process of examination and classification of human chromosome images to diagnose genetic diseases and disorders. It requires time consuming manual examination of cell images by a cytogeneticist to distinguish chromosome classes from each other. Thus, a reliable autonomous human chromosome classification system not only saves time and money but also reduces errors due to the inadequate knowledge level of the expert. Human cell contains 23 pairs of chromosome, 22 autosomes and a pair of sex chromosomes. Hence, we face a multi-class classification task which represents a challenging case for any sort of classifier. In this work, to solve this classification problem, we propose a novel methodology consisting two stages: (i) data preparation and training, and (ii) testing. To determine the most informative content of the dataset several preliminary experiments are conducted and a Principal Component Analysis is done. Then, a single Support Vector Machine (SVM ij ) is trained to separate a pair of classes, (i,j) where a numerical optimization method Pattern Search (PS), is employed to find the optimal parameters for the SVM ij . Considering 22 pairs of autosomes, 22 × 22 experts are trained and optimized. The cluster of experts, we obtain is named as Competitive SVM Teams (CSVMTs) where each SVM ij competes with the others to label a new classification instance. The final output of the classifier is determined by majority voteing. The results obtained on Copenhagen dataset proves the merit of the algorithm as correct classification rates (CRR) on train and test samples are 99.55% and 97.84% respectively, which are higher than any accuracy rate achieved so far in the related literature.

Electrical energy consumption forecasting for Turkey using grey forecasting technics with rolling mechanism

The accuracy of electricity energy consumption prediction is an important issue effecting energy investment decisions as well as environmental policies. Although there are several forecasting techniques, selection of the most accurate technique is vital for energy planners. In this study, grey forecasting techniques with rolling mechanism (RM) have been used for modeling and predicting yearly net electrical energy consumption in Turkey. Three different grey models are generated to find the best model. The best grey model with RM is used for energy consumption forecasting from 2014 to 2030. Furthermore, the effect of RM is studied by comparing the obtained results with a Grey Model without RM. Results show nonhomogeneous Grey Model with RM improves forecasting accuracy.