Yazar "Parlak, Ismail Enes" seçeneğine göre listele

Listeleniyor 1 - 6 / 6
  • Küçük Resim Yok
    Öğe
    Blockchain-assisted explainable decision traces (BAXDT): An approach for transparency and accountability in artificial intelligence systems
    (Elsevier, 2025) Parlak, Ismail Enes
    The increasing opacity and lack of verifiable audit trails in AI decision-making systems pose significant challenges to establishing trust and accountability, particularly in high-impact domains. This paper introduces Blockchain-Assisted Explainable Decision Traces (BAXDT), a novel architecture designed to enhance the transparency and auditability of AI systems. BAXDT creates comprehensive, immutable records for each AI decision by integrating model outputs, SHAP-based XAI summaries, a novel Explanation Density Metric, and detailed model/data context into a unified JSON trace. The 0.80 threshold for the Explanation Density Metric was empirically supported by Kneedle-based automatic threshold detection. The BAXDT architecture leverages blockchain by recording a cryptographic hash of each decision trace on-chain, while the full trace is stored off-chain. The system's effectiveness was demonstrated through a multifaceted evaluation: simulations across three diverse public datasets (medical, financial, educational) confirmed its domain-agnostic applicability; a scalability analysis of up to 20,000 traces demonstrated its efficient and linear performance; and a successful deployment on the Ethereum Sepolia public testnet verified its real-world viability. A case study on text data further underscored the framework's flexibility. BAXDT provides a robust framework for documenting AI decisions-what, why, based on what, and when-thereby fostering trustworthy AI and supporting regulatory compliance.
  • Küçük Resim Yok
    Öğe
    Deep learning-based detection of aluminum casting defects and their types
    (Pergamon-Elsevier Science Ltd, 2023) Parlak, Ismail Enes; Emel, Erdal
    Due to its unique properties, high-pressure aluminum die-casting parts are used quite often, especially in the automotive industry. However, die-casting is a process which requires non-destructive testing of the critical components using technologies such as X-ray to examine the internal defects that are not otherwise visible. Such a timeconsuming visual inspection requires well-trained human specialists with the utmost attention. In this study, state-of-the-art deep learning-based object detection methods were trained using an X-ray image dataset of aluminum parts to detect internal defects and predict their types without human attention. The Al-Cast image dataset used in this study contains 3466 images of parts produced in high-pressure die casting machines. It is shared as an open-access original database for the nondestructive testing (NDT) community. ASTM standard definitions for aluminum casting defects are used in determining their types, and to the best of our knowledge, this novel approach is the first in the deep learning literature. Among the 12 deep learning-based object detection methods used for comparison, YOLOv5 versions yielded the highest detection accuracy (0.956 mAP) with the shortest training time (0.75 h). In addition, tests were performed for both original and contrast enhanced images on 348 test images. YOLOv5m performed an accurate detection performance of 95.9%. Additionally, YOLOv5n can process 132 images per second. This study can be considered the first step of an artificial intelligence product that can detect internal defects of aluminum casting parts with industrial standards and explain the relationship between highpressure injection die casting parameters and these defects.
  • Küçük Resim Yok
    Öğe
    Deep learning-based detection of internal defect types and their grades in high-pressure aluminum castings
    (Elsevier Sci Ltd, 2025) Parlak, Ismail Enes; Emel, Erdal
    With the increasing use of light alloy castings in automobiles, ensuring quality control is essential for safety. Xray imaging offers a practical approach to detecting internal defects in cast components. This study proposes a method to automatically and in real-time identify the location, type, and size of internal defects in aluminum parts produced via high-pressure casting. The proposed two-stage method can detect, segment, and grade defects without expensive hardware in less than a second. Using the YOLOv5 algorithm for defect detection in the first stage, a mean Average Precision (mAP) of 0.971 was achieved. In the second stage, defect grading is performed through segmentation, enabling classification in accordance with international standards without requiring additional training. The methodology provides real-time and highly accurate internal defect quality control and can be applied to different metals and standards. The dataset used in this study contains over 5,000 labelled X-ray images of aluminum cast parts, and it is made available as open access to support the NDT community and researchers.
  • Küçük Resim Yok
    Öğe
    Optimization of the Cutting Parameters Affecting the Turning of AISI 52100 Bearing Steel Using the Box-Behnken Experimental Design Method
    (Mdpi, 2023) Yildiz, Aytac; Ugur, Levent; Parlak, Ismail Enes
    In this study, we aimed to optimize the cutting parameters that affect the minimum temperature and power consumption in the turning of AISI 52100 bearing steel. For this, the Box-Behnken experimental design method, which was used for the lowest number of experiments in the experimental systems created using the response surface method (RSM), was used. The cutting parameters affecting the turning of the AISI 52100 bearing steel were determined as the cutting speed, depth of cut, and feed rate based on a literature research. The temperature and power consumption values were obtained via analyses according to the experimental design method determined by the finite element analysis (FEM) method. The results obtained were analyzed in Design Expert 13 software. According to the analysis results, the parameter values were determined for the minimum temperature and power consumption. The temperature and power consumption variables were affected by all three parameters, namely the cutting speed, depth of cut, and feed rate. For the minimum temperature and power consumption, a cutting speed of 162.427 m/min, depth of cut of 1.395 mm, and feed rate of 0.247 mm/rev, as well as the feed rate parameters, affected both the temperature and power consumption the most. In addition, it was determined that the cutting speed parameter had the least effect on both the temperature and power consumption variables. In addition, validation experiments were carried out in a real experimental environment with optimum values for the cutting parameters. The results showed that the output values obtained within the limits of the study with the obtained equation were quite close (3.3% error for temperature, 6.6% error for power consumption) to the real experimental outputs.
  • Küçük Resim Yok
    Öğe
    THE MOST SUITABLE MOBILE RFID READER SELECTION BY USING INTERVAL TYPE-2 FUZZY TOPSIS METHOD
    (Yildiz Technical Univ, 2018) Yıldız, Aytaç; Karakoyun, Fatih; Parlak, Ismail Enes
    The novelties, innovations and ease of practical applications that the technology brings as it develops have become dispensable parts of our daily life. RFID technology, one of these novelties, forms the basis for new coding, storage and transmission systems; as well as helping businesses to solve the issues that they encounter in controlling data or the troubles faced due to lack of information. RFID applications in supply chain management insure several advantages; providing more accurate and faster communication, ensuring traceability on dealers networks and developing control mechanisms by ensuring real-time information transmission between logistics activities and supply chain management. RFID readers, one of the components of the RFID system, are the cores of RFID systems, and they all have unique and distinct properties that distinguish from each other. Prior to the creation of an RFID system, the selection of the suitable RFID readers that meet all the needs is the leading important issue. In this study, the aim is to select the most suitable mobile RFID reader for a business that aims to organize warehouse transactions in supply chain management and reduce the errors in shipping operations. Since there are multiple criteria in selection of the RFID readers, the problem is considered as a multi-criteria decision making problem and the Interval Type-2 Fuzzy TOPSIS method, which uses type-2 fuzzy sets, is used for solution, as it is more effective in solving the uncertainties. After determining the effective criteria and alternatives in selection of mobile RFID reader, by adopting the steps of the method, "RFID Reader-2" which is superior to other alternatives in terms of "Memory", "Read Distance" and "Drop Spec.", has been chosen as the most suitable mobile RFID reader. Finally, the results that are obtained in the study are evaluated and suggestions for future studies are made.
  • Küçük Resim Yok
    Öğe
    The Structural Behavior of Bridge Arches Under Collapse Load
    (Gazi Univ, 2020) Sakcalı, Gökhan Barış; Gönül, Alper; Parlak, Ismail Enes
    Many of the bridges designed by Mimar Sinan, the leading architect of the Classical Ottoman Period, has continued to exist today thanks to their materials and construction techniques. The reason that arch forms are generally preferred on these bridges is that they are constructed from natural materials with low tensile strength. In the study, 10 bridges and 48 arches in these bridges designed by Mimar Sinan were analysed statistically. Depending on this analysis, 25 arches with varying lengths and heights were modeled by finite element method. For these arches; hinge conditions, collapse loads, maximum displacements, ductility and energy dissipation capacities were examined and regression analyses were performed. Equations have been proposed which gives maximum displacement and collapse load for examined arches.