HEHalime Ergun

Halime Ergun

*This is an algorithmically generated author record based on publicly available information.

Halime Ergun has a total of 10 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • A Genetic Algorithm Model Improved with Zeckendorf Representations for Preventive Maintenance Scheduling Problem

    by Yunus Emre Göktepe, Fikri Köken, Halime Ergun

    This study introduces a novel Genetic Algorithm (GA) model enhanced with Zeckendorf representation and Fibonacci number-based encoding to optimize preventive maintenance scheduling (PMS) problems. Conventional maintenance scheduling methods, based on random or linear encoding techniques, often fail to optimize maintenance processes effectively. Therefore, the proposed model aims to systematically plan maintenance periods and minimize production interruptions by encoding maintenance intervals using Zeckendorf representation. By optimizing maintenance processes, the proposed model enhances system production continuity. Experimental analyses indicate that the proposed model enhances existing production capacity and facilitate a more balanced management of maintenance operations. The electricity and water production capacities increased by 11% and 10%, respectively, while the reserve capacity improved by 9% for electricity and 17% for water. These results show that the proposed method is a new optimization strategy for maintenance planning by enhancing the applicability of GA in preventive maintenance scheduling problems. Optimizing maintenance scheduling with Zeckendorf representation enables systematic and balanced execution of maintenance operations, thereby ensuring more efficient planning in industrial facility maintenance processes.

  • A Novel Genetic Algorithm Model with Fibonacci Encoding and k-Zeckendorf Representations

    by Yunus Emre Göktepe, Fikri Köken, Halime Ergun

    Bu çalışma, karmaşık optimizasyon problemlerini çözmenin verimliliğini artırmak için Fibonacci kodlamasını ve k-Zeckendorf gösterimlerini kullanan yeni bir genetik algoritma (GA) modeli sunmaktadır. Geleneksel GA, kromozomları temsil etmek için ikili veya sayısal kodlamayı kullanır, ancak bu çalışma Fibonacci dizisine ve Zeckendorf teoremine dayanan alternatif bir yaklaşım önermektedir. Bu gösterimleri GA çerçevesine dahil ederek, model arama sürecini iyileştirmeyi ve daha etkili çözümlere yol açmayı amaçlamaktadır. Bu yaklaşımın önemi, GA'ların performansını etkileyen kritik bir faktör olan kromozom gösterimini iyileştirme yeteneğinde yatmaktadır. Yeni kodlama şemaları, algoritmanın keşif ve yararlanma aşamalarını geliştirerek, optimum çözümlere doğru daha verimli bir yakınsama sağlar. Model iki zorlu problemde test edilmiştir: yüksek dereceli bir polinomun optimum değerini bulma ve dikdörtgen prizmanın hacmini optimize etme. Sonuçlar, önerilen yöntemin standart GA'dan daha başarılı olduğunu göstermektedir. Bu araştırma, alternatif kromozom gösterimlerinin GA performansını önemli ölçüde etkileyebileceğini göstermektedir. Ayrıca, özellikle karmaşık polinom denklemleri ve geometrik kısıtlamaları içeren optimizasyon problemleri için yeni bir çerçeve sağlar. GA'ların daha faklı optimizasyon problemlerine uygulanabileceğini gösterir ve hesaplamalı matematik ve mühendislik tasarımındaki potansiyelini vurgular.

  • Open Access

    Modeling Xanthan Gum Foam’s Material Properties Using Machine Learning Methods

    by Halime Ergün, Mehmet Emin Ergün

    Xanthan gum is commonly used in the pharmaceutical, cosmetic, and food industries. However, there have been no studies on utilizing this natural biopolymer as a foam material in the insulation and packaging sectors, which are large markets, or modeling it using an artificial neural network. In this study, foam material production was carried out in an oven using different ratios of cellulose fiber and xanthan gum in a 5% citric acid medium. As a result of the physical and mechanical experiments conducted, it was determined that xanthan gum had a greater impact on the properties of the foam material than cellulose. The densities of the produced foam materials ranged from 49.42 kg/m3 to 172.2 kg/m3. In addition, the compressive and flexural moduli were found to vary between 235.25 KPa and 1257.52 KPa and between 1939.76 KPa and 12,736.39 KPa, respectively. Five machine-learning-based methods (multiple linear regression, support vector machines, artificial neural networks, least squares methods, and generalized regression neural networks) were utilized to analyze the effects of the components used in the foam formulation. These models yielded accurate results without time, material, or cost losses, making the process more efficient. The models predicted the best results for density, compression modulus, and flexural modulus achieved in the experimental tests. The generalized regression neural network model yielded impressive results, with R2 values above 0.97, enabling the acquisition of more quantitative data with fewer experimental results.


  • Open Access

    Box–Behnken experimental design for optimization of chitosan foam materials reinforced with cellulose and zeolite

    by Rıfat Kurt, Halime Ergun, Mehmet Emin Ergun, Abdullah Istek

    Foam materials produced from biopolymers stand out as a more environmentally friendly insulation material solution. This study presents a comprehensive investigation into the development and optimization of chitosan‐based foam materials using a Box–Behnken design. The foams were engineered using varying proportions of chitosan (0.5–3%), cellulose (0.5–3%), and zeolite (0.5–3%), targeting their application as thermal insulators. The physical and thermal properties of the foams that were produced were affected by the type and ratios of components, with density and thermal conductivity ranging from 0.0853 to 0.1915 g cm−3 and 0.0324 to 0.0921 W mK−1, respectively. Higher chitosan content improved insulation properties and mechanical strength whereas zeolite increments increased density and thermal conductivity. Using statistical analysis through the Box–Behnken design, we optimized the foam formulations, achieving minimum thermal conductivity and maximum compression strength at an averaged density, suggesting a strong potential for environmental sustainability applications. The recommended optimal chitosan:cellulose:zeolite composition ratio of 3:3:0.88 provides a valuable insight for tailored foam material formulation. This study shows the relationships between the composition of a composite material and its resultant properties, optimizing its preparation for industrial applicability in an environmentally conscious way within the context of insulation and construction. This investigation contributes to the field of material science by highlighting the versatility and potential of biopolymers but also aligns with the increasing need for green building materials.


  • Open Access

    Wood identification based on macroscopic images using deep and transfer learning approaches

    by Halime Ergun

    Identifying forest types is vital for evaluating the ecological, economic, and social benefits provided by forests, and for protecting, managing, and sustaining them. Although traditionally based on expert observation, recent developments have increased the use of technologies such as artificial intelligence (AI). The use of advanced methods such as deep learning will make forest species recognition faster and easier. In this study, the deep network models RestNet18, GoogLeNet, VGG19, Inceptionv3, MobileNetv2, DenseNet201, InceptionResNetv2, EfficientNet and ShuffleNet, which were pre-trained with ImageNet dataset, were adapted to a new dataset. In this adaptation, transfer learning method is used. These models have different architectures that allow a wide range of performance evaluation. The performance of the model was evaluated by accuracy, recall, precision, F1-score, specificity and Matthews correlation coefficient. ShuffleNet was proposed as a lightweight network model that achieves high performance with low computational power and resource requirements. This model was an efficient model with an accuracy close to other models with customisation. This study reveals that deep network models are an effective tool in the field of forest species recognition. This study makes an important contribution to the conservation and management of forests.


  • INVESTIGATING THE FEASIBILITY OF GUAR GUM-BASED FOAMS FOR INSULATION APPLICATIONS USING REGRESSION ANALYSIS

    by Mehmet Emin Ergün, Halime Ergunb

    Guar gum is commonly utilized in the pharmaceutical, cosmetic, and food industries. However, its use as a foam material for insulation purposes in construction fields has not been extensively studied, especially with regards to machine learning. This study aimed to investigate the potential use of foams produced from biopolymers for insulation and to estimate their properties using two different regression analyses. The foams were produced using a simple and quick procedure involving a mixture of guar gum, cellulose, and boric acid in different proportions, and then dried in the oven. The results of the produced foams showed promising features such as low density, low thermal conductivity, and good mechanical properties, which are highly desirable in insulation materials. A regression model was developed to analyze the effects of the components used in the foam formulation and to provide an estimated method for future research. The regression model was able to accurately predict the results, with an R2 value of up to 0.99, allowing for more quantitative data to be obtained with fewer experimental results. Furthermore, it was found that guar gum had the most significant effect on the properties of the foams. Keywords: Foam, guar gum, thermal conductivity, regression, insulation


  • Open Access

    Hikayeler İçin Artırılmış Gerçeklik Yaklaşımı

    by Yusuf Uzun, Halime Ergün, Esfanur Şeker

    Son yıllarda gelişen teknoloji ile birlikte yaşam biçimimiz ve alışkanlıklarımız da değişmektedir. Her geçen gün birçok yeni uygulama ortaya çıkmaktadır. Bu uygulamalardan biri de artırılmış gerçeklik uygulamasıdır. Artırılmış gerçeklik çevremizde algıladığımız fiziksel unsurların üzerinde bilgisayar kaynaklı grafik, video ve ses gibi verilerin gerçek zamanlı olarak görüntüleyen gerçek dünya ile sanal dünyanın iç içe girdiği bir teknolojidir. Askeri, sanayi, tıbbi, ticari, reklam ve eğlence alanlarında sıklıkla kullanılan artırılmış gerçeklik eğitim alanında da kullanılmaya başlanmıştır. Eğitim alanında kullanılmasındaki amaç çocukların merak duygusunu geliştirmek, dersleri eğlenceli hale getirerek öğrenme sürecini kolaylaştırmak ve sınıf ortamında yapılması mümkün olmayan konuların hazırlanmasına olanak sağlamasıdır. Bu çalışmada, hikâye kitaplarındaki heyecanlı kısımları çocuklara hissettirmek, konuya dikkatlerini çekmek ve okuma isteklerini artırmak için artırılmış gerçeklik teknolojisi kullanılmıştır. Öğrenciler üzerinde gerçekleştirilen testlerde başarılı dönüşler elde edilmiştir.


  • DETERMINATION OF HOMOGENE RAYS WITH MORPHOLOGICAL PROCESSES IN HARDWOOD AND SOFTWOOD

    by Halime Ergün

    Tr en Bu çalışmada, yapraklı ve iğne yapraklı ağaçlarda (Abies alba Mill, Tilia platyphyllos Scop., Tilia cordata Mill, Betula alba, Juglans regia L. Walnut, Ulmus scabra Mill) bulunan homojen öz ışınları görüntü işleme metotları kullanılarak belirlenmiştir. Öz ışınları paranşim hücrelerin bir araya gelmesiyle oluşur ve teğet kesitte gözenekli bir yapıya sahiptir. Küçük paranşim hücrelerinden oluştuğu için eşikleme işleminden sonra büyük bölgeler görüntüden çıkarılmıştır. Yükseklik/genişlik oranı fazla olan bölgeler de görüntüden silinmiştir. Morfolojik işlemlerden biri olan kapama işlemi ile paranşim hücreleri birleştirilerek öz ışınları bulunmuştur. Birleşmeyen ve öz ışını olamayan bölgeler silinmiştir. Genişleme ve doldurma işlemi ile öz ışınların son şekli belirlenmiştir. Elde edilen sonuçlar görsel ve istatistiksel olarak verilmiştir. Yapılan analiz sonucunda; en fazla ve en az öz ışını sırasıyla Ulmus scabra Mill. ve Abies alba Mill. de tespit edilmiştir. Belirlenmek istenen yapıya uygun görüntü işleme tekniklerinden faydalanılarak öz ışınları otomatik olarak bulan sistemler, odun anatomisi çalışmalarının kısa sürede ve daha kolay bir şekilde yapılmasını sağlamaktadır. In this study, homogeneous rays were determined by using image processing methods in hardwood and softwood (Abies alba Mill, Tilia platyphyllos Scop., Tilia cordata Mill, Betula alba, Juglans regia L. Walnut, Ulmus scabra Mill). The rays that parenchyma cells are come together, have a porous structure in tangential section. Since the rays are composed of small parenchyma cells, large regions were removed from the image after thresholding. The regions that high/width ratio was high were deleted from the image. One of the morphological processes, the close process was combined with the parenchyma cells after rays were determined. Non-combining and non-rays regions were deleted. The expansion and filling process determined the final shape of the rays. The results were given statistically and visually. As a result of the analysis; maximum and minimum rays were determined Ulmus scabra Mill. and Abies alba Mill. respectively. Systems that find the rays automatically by utilizing image-processing techniques appropriate to the structure to be determined ensure that the study of wood anatomy is done in a short time and more easily.