publications

2023

1. Distribution Center Location Selection in Humanitarian Logistics Using Hybrid BWM–ARAS: A Case Study in Türkiye

   Erden, Caner, Ateş, Çağdaş,  and Esen, Sinan

   Journal of Homeland Security and Emergency Management Jun 2023

   Abs DOI

   This study investigates the criteria affecting the location of humanitarian logistics distribution centers in the Sakarya province of Turkey, an area prone to natural disasters. The study identifies potential distribution center locations and uses the BestWorst Method (BWM) to determine criteria such as population, distance to major highways and airports, public transportation availability, natural disaster risk, and suitable infrastructure. BWM is used to assign weights to each criterion and rank them based on their importance. The Additive Ratio Assessment (ARAS) method is then used to evaluate potential distribution center locations based on the established criteria. Disaster management experts and academicians provide their opinions through an online and face-to-face survey. Based on the results, Adapazarı is identified as the most suitable district for a humanitarian logistics distribution center. The study highlights the importance of considering multiple criteria when selecting distribution center locations and provides a framework for using multi-criteria decision-making methods in logistics planning. Disaster managers and policymakers can use the results to make informed decisions about the location of humanitarian logistics distribution centers.

2. Integrated Process Planning, Scheduling, and Due-Date Assignment

   Demir, Halil Ibrahim, Kökçam, Abdullah Hulusi,  and Erden, Caner

   Jun 2023

   Abs DOI

   Traditionally, the three most important manufacturing functions are process planning, scheduling, and due-date assignment, which are handled sequentially and separately.This book integrates these manufacturing processes and functions to increase global performance along with manufacturing and production cost savings. Integrated Process Planning, Scheduling, and Due-Date Assignment combines the most important manufacturing functions to use manufacturing resources better, reduce production costs, and eliminate bottlenecks with increased production efficiency. The book covers how the integration will help eliminate scheduling conflicts and how to adapt to irregular shop floor disturbances. It also explains how other elements, such as tardiness and earliness, are penalized and how prioritizing helps improve weight performance function. This book will draw the interest of professionals, students, and academicians in process planning, scheduling, and due-date assignment. It could also be supplemental material for manufacturing courses in industrial engineering and manufacturing engineering departments.

3. Enhancing Machine Learning Model Performance with Hyper Parameter Optimization: A Comparative Study

   Erden, Caner, Demir, Halil Ibrahim,  and Kökçam, Abdullah Hulusi

   Feb 2023

   Abs DOI

   One of the most critical issues in machine learning is the selection of appropriate hyper parameters for training models. Machine learning models may be able to reach the best training performance and may increase the ability to generalize using hyper parameter optimization (HPO) techniques. HPO is a popular topic that artificial intelligence studies have focused on recently and has attracted increasing interest. While the traditional methods developed for HPO include exhaustive search, grid search, random search, and Bayesian optimization; meta-heuristic algorithms are also employed as more advanced methods. Meta-heuristic algorithms search for the solution space where the solutions converge to the best combination to solve a specific problem. These algorithms test various scenarios and evaluate the results to select the best-performing combinations. In this study, classical methods, such as grid, random search and Bayesian optimization, and population-based algorithms, such as genetic algorithms and particle swarm optimization, are discussed in terms of the HPO. The use of related search algorithms is explained together with Python programming codes developed on packages such as Scikit-learn, Sklearn Genetic, and Optuna. The performance of the search algorithms is compared on a sample data set, and according to the results, the particle swarm optimization algorithm has outperformed the other algorithms.

4. Genetic Algorithm-Based Hyperparameter Optimization of Deep Learning Models for PM2.5 Time-Series Prediction

   Erden, C.

   Int. J. Environ. Sci. Technol. Mar 2023

   Abs DOI

   Since air pollution negatively affects human health and causes serious diseases, accurate air pollution prediction is essential regarding environmental sustainability. Although conventional statistical and machine learning methods have been widely used for air quality forecasting, they have limitations in finding nonlinear relations and modeling sequential data. In recent years, deep learning methods such as long short-term memory, recurrent neural networks, and gated recurrent units have been successfully applied in several research areas, including time-series forecasting. In this study, deep learning algorithm is employed to predict the PM2.5 dataset, including air pollutants (NO, NO2, NOX, O3, PM2.5, SO, and SO2) and meteorological features (wind speed, wind direction, and air temperature) in Istanbul metropolitan. Deep learning algorithms have many hyperparameters such as learning and dropout rate, the number of hidden layers and units in each hidden layer, activation function, loss function, and optimizer that need to be optimized in order to achieve optimal training performance. Therefore, a genetic algorithm-based hyperparameter optimization approach is proposed to find the best parameter combination. The prediction results of deep learning algorithms are compared with default hyperparameters and random search algorithms to confirm the efficacy of the genetic algorithm approach. The proposed method outperforms the other configurations, with the MSE error reduced by 13.38% and 55.30% for testing performance, respectively. The experimental results revealed that genetic algorithms are promising and applicable in hyperparameter optimization of deep neural network models, especially in air quality forecasting.

5. A Modified Integer and Categorical PSO Algorithm for Solving Integrated Process Planning, Dynamic Scheduling and Due Date Assignment Problem

   Erden, Caner, Demir, Halil Ibrahim,  and Canpolat, Onur

   Scientia Iranica Mar 2023

   PDF DOI
6. Predicting next Hour Fine Particulate Matter (PM2.5) in the Istanbul Metropolitan City Using Deep Learning Algorithms with Time Windowing Strategy

   Eren, Beytullah, Aksangür, İpek,  and Erden, Caner

   Urban Climate Mar 2023

   Abs DOI

   Poor air quality has various detrimental physical and mental effects on human health and quality of life. In particular, PM2.5 air pollution has been associated with cardiovascular and respiratory problems. Therefore, air quality management is an essential issue for densely populated cities to reduce or prevent the adverse effects of air pollution. Considering this, reliable models for predicting pollution levels for pollutants like PM2.5 are critical tools for decision-making. For this purpose, this study presents three kinds of deep learning (DL) algorithms (LSTM, RNN, and GRU) that utilize a time-windowing strategy to predict the hourly concentration of PM2.5 in the Istanbul metropolitan. The models were trained and tested using large data sets that envelope air quality parameters (PM2.5, SO2, NO, NO2, NOX, and O3) and meteorological factors (temperature, wind speed, relative humidity, and air pressure) for about five years. The experimental results demonstrate that the LSTM+LSTM model performs significantly better with an R2 of 0.98 and 0.97 at the significance level (p < 0.05) for training and test sets compared to other deep learning algorithms. In addition, data for one year from several stations located in nine different districts of Istanbul were used to evaluate the proposed model’s generalization ability. As a result, the proposed LSTM+LSTM model has a good generalization ability with an R2 accuracy rate of 0.90 (p < 0.05) and above for all stations and can be used for non-linear, non-stationary multidimensional time series data. Furthermore, the results were compared to other studies in the literature; it was found that the proposed LSTM+LSTM model performed better in predicting PM2.5 concentrations.

7. Ant Colony Optimization Application in Bottleneck Station Scheduling

   Kılıçaslan, Emre, Demir, Halil Ibrahim,  Kökçam, Abdullah Hulusi and 2 more authors

   Advanced Engineering Informatics Apr 2023

   Abs DOI

   Finding optimal solutions to production planning and scheduling problems is crucial for surviving in a competitive environment and meeting customer expectations over time. Planning can become complicated in sectors with many different products such as tire production. This study focuses on the bottleneck problem caused by a machine called a Quadruplex Extruder in a tire factory. With this machine, rubber is extruded and transformed into a tread material product, which is critically important in some essential tire features, such as low rolling resistance and brake distance. This study aims to minimize the set-up times in production by optimizing the manufacturing order of the products produced in a quadruplex extruder machine using the Ant Colony Algorithm (ACA), a well-known metaheuristic method to solve polynomial optimization problems. In addition, the second version of the Lin–Kernighan–Helsgaun (LKH-2) algorithm was adapted to this problem. Manually prepared, LKH-2 and ACA-produced schedules were compared in terms of global efficiency. As a result, it has been shown that ACA can provide fast and suitable solutions for decision makers in production planning.

8. A Hyper Parameterized Artificial Neural Network Approach for Prediction of the Factor of Safety against Liquefaction

   Kurnaz, Talas Fikret, Erden, Caner,  Kökçam, Abdullah Hulusi and 2 more authors

   Engineering Geology Jun 2023

   Abs DOI

   Soil liquefaction during earthquakes is a complex geotechnical engineering problem. Although various analytical approaches exist for predicting liquefaction risk, their limitations have led researchers to explore using artificial intelligence and machine learning methods. Machine learning has the potential to significantly improve the ability to predict soil liquefaction and mitigate the associated risks. This study proposes a hyper parameterized artificial neural network architecture using random search, grid search, and Bayesian optimization algorithms to predict the factor of safety against liquefaction. The performances of hyper parameterized machine learning algorithms, including artificial neural networks (ANN), decision trees (DT), random forest (RF), and support vector regression (SVR), were compared. Statistical tests show that the proposed ANN outperformed the others with a determination coefficient of 0.99 at a 95% significance level. Hyperparameter optimization significantly improved learning performance with up to a 48% reduction in RMSE scores. The proposed method was compared with previous studies, and performance results confirmed its effectiveness and generalization ability. In conclusion, this study highlights the potential of machine learning algorithms for predicting soil liquefaction and emphasizes the importance of hyperparameter optimization for improving model performance. The findings of this study have practical implications for improving liquefaction risk assessment and mitigating the associated hazards.

9. Machine Learning Experiment Management with MLFlow

   Erden, Caner

   Jun 2023

   Abs

   Big data and machine learning are driving the Fourth Industrial Revolution. With the age of big data upon us, we risk drowning in a flood of digital data. Big data has now become a critical part of both the business world and daily life, as the synthesis and synergy of machine learning and big data...

10. Derin Öğrenme ve ARIMA Yöntemlerinin Tahmin Performanslarının Kıyaslanması: Bir Borsa İstanbul Hissesi Örneği

    Erden, Caner

    Yönetim ve Ekonomi Dergisi Sep 2023

    Abs DOI

    Finansal zaman serisi verileri doğrusal olmayan, karmaşık, birçok ekonomik faktörden etkilenen ve tahmin edilmesi zor verilerdir. Çok boyutlu ilişkilerin tahminini gerektiren finansal zaman serisi modelleri için çeşitli istatistiksel yöntemler geliştirilmiştir. Ancak günümüzde büyük verilerin kaydedilmesi, analiz edilmesi ve anlamlı bilgiye dönüştürülmesi kolaylaştığından dolayı finansal tahmin geliştirmede makine öğrenmesi algoritmalarının kullanımı özellikle son yıllarda artmıştır. Bu çalışmada, Borsa İstanbul endeksinde metal ana pazarında işlem gören EREGL hissesine ait veriler zaman serisi yöntemleri ile analiz edilmiş ardından ARIMA ve derin öğrenme modelleri ile tahmin edilmiştir. Geliştirilen derin öğrenme yönteminde veri ön işleme aşamaları, özellik çıkarımı çalışmaları ve farklı zaman çerçeveleri ile tahmin performansı iyileştirilmiştir. Derin öğrenme algoritmalarının zaman serisi çalışmalarında kullanılabilmesi için zaman gecikmelerinden oluşan bir çerçeve kullanılmalıdır. Bu çalışmada, farklı zaman gecikmeleri için senaryolar denenmiş ve performans kıyaslaması ARIMA modelleri ve uzun-kısa vadeli bellek (LSTM), geçitli tekrarlayan ünite (GRU) ve özyineli sinir ağları (RNN) algoritmalarını kullanan derin öğrenme modelleri arasında gerçekleştirilmiştir. Deneysel çalıştırmalar ile RNN algoritmasının diğerlerine göre daha iyi tahmin performansına sahip olduğu ve ele alınan test veri seti üzerinde ortalama %93’lük doğrulukla tahmin ettiği ortaya konulmuştur.

2022

1. Evaluation of Data Preprocessing and Feature Selection Process for Prediction of Hourly PM10 Concentration Using Long Short-Term Memory Models

   Aksangür, İpek, Eren, Beytullah,  and Erden, Caner

   Environmental Pollution Oct 2022

   Abs DOI

   Studies have confirmed that PM10, defined as respirable particles with diameters of 10 μm and smaller, has adverse effects on human health and the environment. Various estimation methods are employed to determine the PM10 concentration using historical data on controlling PM10 air pollution, early warning, and protecting public health and the environment. The present study analyses different Long Short-Term Memory (LSTM) models that can predict hourly PM10 concentration. In parallel, the study also investigates the effectiveness of the data preprocessing and feature selection (DPFS) process on the prediction accuracy of the LSTM models. For this purpose, three different LSTM models, namely Vanilla, Bi-Directional, and Stacked, were developed. Then, a comprehensive data preprocessing stage is used to eliminate missing and erroneous data and outliers from real-world raw data, and a feature selection process is applied to extract unnecessary features. The LSTM models consider three air quality parameters, including SO2, O3, and CO, and three meteorological factors, including relative humidity, wind direction, and wind speed. The prediction performances of the LSTM models are compared using the RMSE, MAE and R2 performance index according to whether DPFS is used in the models or not. As a result, when the DPFS process was applied, the proposed LSTM models achieved high prediction performance and can be used to predict hourly PM10 concentrations. Overall, the DPFS process significantly enhanced the developed LSTM models’ prediction performance. Furthermore, the proposed model might be a useful tool for city administrators to make decisions and improve air quality management efforts.

2. The Transition to 5G Project In The Telecommunications Industry Using CPM and PERT Techniques

   Arslan, Faruk, Demir, Halil Ibrahim,  Erden, Caner and 1 more author

   In 4th International Conference on Applied Engineering and Natural Sciences Nov 2022
3. Taş\ima Modları ve Ulaşt\irma Modelleri

   Erden, Caner

   Nov 2022
4. Makine Öğrenmesinde Hiper Parametre Optimizasyonu

   Erden, Caner

   Nov 2022

   Abs

   Yapay zeka ve makine öğrenmesi uygulamaları iş dünyası ndan akademiye, sağlı k alanı ndan eğitim sektörüne kadar hemen hemen tüm alanlardaki etkisini hı zlı bir şekilde artı rmaktadı r. Makine öğrenmesi kullanan organizasyonlar kullanmayan organizasyonlara göre daha başarı lı uygulamalar gerçekleştirdiğinden dolayı makine öğrenmesi uygulamaları geçmişe kı yasla daha kolay ve küçük işletmeler tarafı ndan bile uygulanabilir hale gelmiştir. Ancak verimli çalışan ve sürekliliği olan makine öğrenmesi uygulamaları üretmek zor ve karmaşı k bir süreçtir. Geçmişteki verilerden faydalanarak çalışan makine öğrenmesi uygulamaları nı n yeni veri girişine adapte edilmesi, öğrenmenin sürekliliğinin sağlanması gibi zorluklar aynı zamanda günümüz makine öğrenmesi uygulamaları nı öne çı karan özelliklerdendir. Bu anlamda, makine öğrenmesi alanı ndaki en önemli konulardan birisi de Hiper Parametre Optimizasyonu(HPO) konusudur. İyi gerçekleştirilmiş bir HPO ile gerçeğe daha uygun performansa sahip makine öğrenmesi çalışmaları geliştirilebilir. Bu bölümde ilk olarak HPO konusunda yapı lan çalışmalar özetlenerek HPO konusuna duyulan ihtiyaçtan bahsedilecektir. İkinci kı sı mda, HPO çözümü için geliştirilmiş yöntemler ve çözüm araçları anlatı lacaktı r. Son olarak, uygulamalarda sı klı kla kullanı lan ve güçlü bir HPO aracı olan Hyperopt aracı ndan bahsedilecek ve bir uygulama üzerinde aracı n kullanı mı gösterilecektir.

5. Machine Learning Approaches for Prediction of Fine-Grained Soils Liquefaction

   Ozsagir, Mustafa, Erden, Caner,  Bol, Ertan and 2 more authors

   Computers and Geotechnics Dec 2022

   Abs DOI

   Since soil liquefaction is a dimension that increases the amount and severity of losses in an earthquake, it is vital to estimate the liquefaction potential accurately. Traditionally, several analytical inferences were made for the prediction of soil liquefaction. However, it is necessary to use machine learning methods to establish nonlinear relationships of soil physical characteristics and develop an accurate classification model. In this study, the applicability of seven different machine learning algorithms; decision trees, logistic regression, support vector machines, k-nearest neighbors, stochastic gradient descent, random forest, and artificial neural network, were investigated on a data set obtained from field experiments (Standard Penetration Test) on soils in Adapazari region after the 1999 earthquake. Performance metrics such as accuracy, recall, precision, F1 score, and receiver operating characteristic evaluated algorithms. As a result of experimental studies, the decision tree algorithm performed best on the dataset, with an overall accuracy of 90%. The decision tree model provides an easy and effective tool for evaluating ground liquefaction potential to decision-makers. As a result of the decision tree study, it was observed that the mean grain size (D50) soil feature has the most significant effect on the liquefaction potential.

6. Recent Advances in Intelligent Manufacturing and Service Systems: Select Proceedings of IMSS 2021

   Dec 2022
7. Economic Analysis and Determination of Profitability of an Automotive Company’s Building Construction Investment

   Yuvalı, Enes, Demir, Halil Ibrahim,  Kökçam, Abdullah and 1 more author

   In 4th International Conference on Applied Engineering and Natural Sciences Nov 2022

2021

1. A Multi-Criteria Decision Making and Goal Programming Model for Fire Station Location Selection

   Alkan, Melek, Demir, Halil,  Erden, Caner and 1 more author

   In Proceedings of 11th International Symposium on Intelligent Manufacturing and Service Systems May 2021
2. Alternative Cleaning Company Selection Method With Fuzzy Topsis

   Bayındır, Sümeyye, Demir, Halil,  Erden, Caner and 1 more author

   In Proceedings of 11th International Symposium on Intelligent Manufacturing and Service Systems May 2021
3. Hybrid Evolutionary Strategy and Simulated Annealing Algorithms for Integrated Process Planning, Scheduling and Due-Date Assignment Problem

   Demir, Halil I\.ḃrahim, Phanden, Rakesh,  Kökçam, Abdullah and 2 more authors

   Academic Platform Journal of Engineering and Science Jan 2021

   DOI
4. A Tabu Search and Hybrid Evolutionary Strategies Algorithms for the Integrated Process Planning and Scheduling with Due-date Agreement

   Demi̇R, Halil İbrahim, Erden, Caner,  Kökçam, Abdullah and 1 more author

   Journal of Intelligent Systems: Theory and Applications Mar 2021

   DOI
5. Python İle Veri Madenciliği

   Erden, Caner

   Mar 2021
6. Green Supply Chain Management in the Context of Sustainability

   Koç, Serkan,  and Erden, Caner

   Journal of Business and Trade Mar 2021

   Abs

   For centuries, humanity has seen nature as an unlimited resource, despised it, they polluted it. Hence, that caused environmental problems. On the one hand, the rapidly increasing population phenomenon, the existence of natural resources that are running out has made it necessary to seek new solutions for humanity. The solution put forward in this framework is the understanding of sustainable development, which can be summarized as ensuring that natural resources are transferred to future generations without completely consuming them. The priority in sustainable development approach has been the sustainability of the natural environment. With the increasing awareness of the international community and states, production systems and processes are revised and transition to green practices that will minimize environmental damage has rapidly gained importance. A traditional supply chain is a one-way process that involves productionrelated activities from raw material procurement to delivery of the final product. Due to environmental requirements affecting production operations today, an increasing emphasis is placed on the development of environmental management strategies for the supply chain. In this study, green approaches, and green supply chain management (GSCM) are considered conceptually, and the green activities implemented and the obstacles in front of them are evaluated. A general perspective is put forward to ensure and maintain the green supply chain.

2020

1. Dynamic Integrated Process Planning, Scheduling and Due-Date Assignment Using Ant Colony Optimization

   Demir, Halil Ibrahim,  and Erden, Caner

   Computers & Industrial Engineering Nov 2020

   DOI

2019

1. Concurrent Solution of WATC Scheduling with WPPW Due Date Assignment for Environmentally Weighted Customers, Jobs and Services Using SA and Its Hybrid

   Demir, Halil Ibrahim, Erden, Caner,  Kökçam, Abdullah Hulusi and 1 more author

   PEN Jan 2019

   DOI
2. Solving Integrated Process Planning, Dynamic Scheduling, and Due Date Assignment Using Metaheuristic Algorithms

   Erden, Caner, Demir, Halil Ibrahim,  and Kökçam, Abdullah Hulusi

   Mathematical Problems in Engineering May 2019

   Abs DOI

   Because the alternative process plans have significant contributions to the production efficiency of a manufacturing system, researchers have studied the integration of manufacturing functions, which can be divided into two groups, namely, integrated process planning and scheduling (IPPS) and scheduling with due date assignment (SWDDA). Although IPPS and SWDDA are well-known and solved problems in the literature, there are limited works on integration of process planning, scheduling, and due date assignment (IPPSDDA). In this study, due date assignment function was added to IPPS in a dynamic manufacturing environment. And the studied problem was introduced as dynamic integrated process planning, scheduling, and due date assignment (DIPPSDDA). The objective function of DIPPSDDA is to minimize earliness and tardiness (E/T) and determine due dates for each job. Furthermore, four different pure metaheuristic algorithms which are genetic algorithm (GA), tabu algorithm (TA), simulated annealing (SA), and their hybrid (combination) algorithms GA/SA and GA/TA have been developed to facilitate and optimize DIPPSDDA on the 8 different sized shop floors. The performance comparisons of the algorithms for each shop floor have been given to show the efficiency and effectiveness of the algorithms used. In conclusion, computational results show that the proposed combination algorithms are competitive, give better results than pure metaheuristics, and can effectively generate good solutions for DIPPSDDA problems.

2018

1. Process Planning and Scheduling with WNOPPT Weighted Due-Date Assignment Where Earliness, Tardiness and Due-Dates Are Penalized

   Demi̇r, Halil İbrahim, Canpolat, Onur,  Erden, Caner and 1 more author

   Journal of Intelligent Systems: Theory and Applications Sep 2018

   DOI
2. Solving Process Planning, ATC Scheduling and Due-date Assignment Problems Concurrently Using Genetic Algorithm for Weighted Customers

   Erden, Caner, Demir, Halil İbrahim,  Göksu, Alper and 1 more author

   Academic Platform-Journal of Engineering and Science Jan 2018

   DOI

2017

1. Integrating Process Planning, WATC Weighted Scheduling, and WPPW Weighted Due-Date Assignment Using Pure and Hybrid Metaheuristics for Weighted Jobs

   Demir, Halil Ibrahim, Erden, Caner,  Demiriz, Ayhan and 2 more authors

   International Journal of Computational and Experimental Science and Engineering Jun 2017

   DOI
2. Proses Planlama ve Ağı rlı klı Teslim Tarihi Atama Ile Birlikte Ağı rlı klı Çizelgeleme Probleminin Bazı Saf ve Melez Meta-Sezgisel Yöntemler Ile Çözümü

   Demir, Halil İbrahim,  and Erden, Caner

   SAÜ Fen Bilimleri Enstitüsü Dergisi Apr 2017

   DOI

2016

1. Accident Causation Factor Analysis of Traffic Accidents using Rough Relational Analysis:

   Erden, Caner,  and Çelebi, Numan

   International Journal of Rough Sets and Data Analysis Jul 2016

   Abs DOI

   The aim of this study is to show that the decision rules generated from Rough Sets Theory can be used for a new relational analysis. Rough Sets Theory generally works with small datasets more than big data. If we can deal with the decision rules and its complexities, it is still possible to analyze big data with Rough Set Theory. That is why in this study the authors offer a statistical method to overdue problems which belongs to big data. According statistical methods, a lots of decision rules generated from rough sets theory become useful information. Using a real case data on the traffic accident which were taken place in USA in 2013, this paper finds the relationships between accident causation factors which may be referred to decision makers in the field of traffic.

2. İplik Sürtünme Özelliklerinin İncelemesinde Kaba Kümeler Yaklaşı mı

   Erden, Caner,  and Nazarov, Muhammed

   Erzincan University Journal of Science and Technology Jun 2016

   DOI

2015

1. Gri Sistem Teorisi Kullanı larak Türkiye’nin Büyüme Oranı Faktörlerinin Analizi

   Ceviz, Emre,  and Erden, Caner

   SAUFenBilDer Aug 2015

   DOI