Kürşat Yildiz

Autonomous vehicle technologies are reshaping traffic safety and urban mobility; however, software and sensor failures, legal uncertainties, and mixed traffic environments continue to pose significant accident risks. Recent fatal incidents during real-world testing underline the critical need for effective risk prediction approaches. While existing studies mainly focus on accident causes and probabilities, deep learning research is often limited to in-vehicle sensor data.This study proposes a hybrid framework integrating machine learning and deep learning models with Geographic Information Systems (GIS)-based spatial analysis to predict both the occurrence and severity of autonomous vehicle accidents and to identify high-risk locations and conditions. The results show that Support Vector Machine (SVM) and Artificial Neural Network (ANN) models perform effectively in accident severity prediction, while ANN achieves superior performance in accident occurrence prediction under balanced data conditions. The proposed approach offers a data-driven basis for risk-aware route planning and dynamic safety management in intelligent transportation systems.

This study develops a mixed‐integer nonlinear optimisation model to jointly minimise PM ₁₀ and CO ₂ emissions from urban road transport in nine of Türkiye's most densely populated provinces, aligning with the targets of the Paris Agreement. The model is novel in its integration of environmental baselines (city‐level PM ₁₀ and CO ₂ levels), economic feasibility (GDP), and urban form variables (population density and land‐use ratios), creating a realistic and context‐aware decision framework. It applies three tiers of constraints: (i) 2023 regulatory emission ceilings, (ii) GDP‐based economic capacity thresholds, and (iii) nonlinear constraints reflecting how urban morphology influences policy adoption. A composite climate impact index—capturing post‐policy PM ₁₀ , CO ₂ , and their interaction—serves as the objective function to yield granular, city‐specific mitigation scenarios. Results indicate that cities like Ankara, Bursa, and Antalya can virtually eliminate PM ₁₀ emissions, while Istanbul achieves the highest CO ₂ reduction (87%), collectively delivering nearly 20 Tg CO ₂ ‐equivalent reduction. Sensitivity analysis confirms that moderate increases in GDP can enhance mitigation outcomes by expanding feasible policy space. As the first model in the Turkish context to embed both urban planning and economic structure into dual‐pollutant optimisation, this framework offers not only predictive capacity but also prescriptive guidance for policymakers. It enables comparative assessment of emission‐reduction strategies—such as fleet electrification, low‐emission zones, and transit infrastructure investment—offering a robust and transferable tool for other rapidly urbanising regions aiming to simultaneously advance air quality, public health, and climate resilience.

This study presents a machine-learning framework that predicts a fracture-based healing index of asphalt mixtures by explicitly incorporating image-quantified fracture-surface damage modes (adhesive, cohesive, aggregate). Damage types were quantified through digital image processing. Two datasets were employed: one with specimens broken at –20 °C and another with variable temperatures (–20 °C to 20 °C). Eight feature sets were developed to isolate key factors, and multiple ML models were tested. Results showed that breaking temperature is the most dominant factor influencing healing, though its strong correlation can create spurious relationships that mask the effects of mixture properties. When temperature was fixed, aggregate damage consistently emerged as the most reliable predictor, with the best performance achieved by Support Vector Regressor (R² = 0.856 at –20 °C). Bitumen content showed gradation-dependent effects: in porous mixtures, higher binder reduced aggregate damage, while in dense mixtures the effect was negligible. Regardless of gradation, higher binder content enhanced healing by improving crack filling and binder flow. Air voids also showed contrasting effects: healing decreased with higher voids in dense mixtures, but moderate voids in porous mixtures facilitated binder redistribution and improved healing. Among the algorithms, Support Vector Regressor achieved the highest predictive accuracy, followed by Gradient Boosting, while Linear Regression underperformed, reflecting the nonlinear nature of healing. Feature selection with Recursive Feature Elimination and Cross-Validation (RFECV) improved efficiency with minor accuracy loss, though excluding aggregate damage reduced reliability. Sensitivity analyses confirmed that breaking temperature dominated predictions at variable conditions, while at fixed temperature, volumetric properties and cohesive damage became more influential. These findings demonstrate the potential of ML to capture complex healing mechanisms and support mix design strategies tailored to gradation type and service temperature.

This research explores the feasibility of incorporating a 1:3 lime and brick dust blend—derived from the core components of traditional Khorasan mortar—as a modifier in hot mix asphalt (HMA). The additive was introduced into the conventional aggregate-bitumen mixture by replacing fine and filler aggregates at a rate of 2% of the total aggregate mass. Its influence on the performance of the wearing course in road pavements was analyzed through a comprehensive set of laboratory experiments. These tests assessed key parameters, including air void content (Pa), optimum bitumen ratio, voids in mineral aggregate (VMA), voids filled with asphalt (VFA), bulk specific gravity (Gmb), Marshall stability and flow, indirect tensile strength, and wheel rutting depth, to evaluate durability, flexibility, and resistance to deformation. The findings indicated that the modified mixture outperformed conventional blends, significantly improving the long-term durability of the pavement surface. Furthermore, considering the reuse potential of industrial waste bricks, the lime-brick dust mixture emerged as an environmentally sustainable and viable additive for road pavement applications.

This research introduces the Urban Traffic Mobility Optimization Model (UTMOM), a data-driven methodology for analyzing two distinctive urban traffic datasets through the integration of data mining and mathematical modeling. Designed to decode the complexities of urban mobility patterns, UTMOM meticulously evaluates daily traffic dynamics with a focus on reducing discrepancies and underscoring variations in traffic intensity, particularly during peak times. Our findings unveil pivotal insights into the differences across datasets, providing a substantial contribution to the realms of traffic management and urban planning. UTMOM delves into the intricacies of traffic flow variations, emphasizing the critical importance of comprehending fluctuations in traffic volume across diverse times and locations. By incorporating detailed graphical representations and statistical validations, including ANOVA analysis, our study delivers a comprehensive evaluation of UTMOM’s precision in reflecting real-world traffic scenarios. These insights affirm the value of data-informed strategies in optimizing traffic flow and alleviating congestion. Positioned as a valuable asset for traffic engineers, data scientists, and urban planners, UTMOM advocates for advanced modeling techniques to improve urban mobility. Beyond enriching academic discourse on traffic analysis, UTMOM offers actionable intelligence for enhancing the efficiency and sustainability of urban transportation systems. Through this in-depth investigation, our aim is to catalyze the development of innovative solutions to traffic challenges, steering towards smoother and more sustainable urban environments.

This study underscores the importance of prioritizing transportation modes in Ankara, particularly given the pivotal role transportation holds in contemporary urban societies. Transportation directly shapes the socio-economic framework of metropolitan areas. To address the complexities of transportation in Ankara, the study introduces a hybrid approach by integrating the Fuzzy CRITIC (Criteria Importance Through Inter-Criteria Correlation) method with the novel Pythagorean Fuzzy Weighted Sum Method. This novel approach assesses the various transportation modes available in Ankara, taking into account pivotal criteria such as cost, duration, reliability, comfort, and flexibility. The hybridized methodology offers a systematic way to determine the weights of each criterion. Then, leveraging these weights, the performance of each transportation mode is calculated and ranked. This integrated approach proves to be a powerful analytical tool for addressing multi-criteria decision-making challenges, especially when confronted with uncertainty and intricate details. The outcomes of this research aim to serve as a cornerstone for the Ankara Ministry of Transportation and other key stakeholders. The insights derived can be pivotal for enhancing the existing transportation infrastructure or for the initiation of new, more efficient projects. Beyond its direct applications, this study emphasizes the versatility and robustness of hybrid multi-criteria decision-making methods, positioning itself as a benchmark for analogous urban transportation challenges. In essence, this research proposes an advanced, scientific strategy to streamline transportation modes in Ankara. It recognizes the intertwined complexities of urban transport and offers a comprehensive analytical approach that can drive transformative decisions in the city's transportation landscape.

Araştırmalar, üretim ve envanter yönetimi alanında itme-çekme stratejilerinin büyük öneme sahip olduğunu ve lojistik süreçlerde kritik bir rol oynadığını göstermektedir. Üretim sürecinde zaman ve kaynak tasarrufu sağlamak amacıyla çeşitli itme-çekme stratejileri geliştirilmiştir. Şirketleri etkileyen en önemli kriterleri belirleyerek, en uygun stratejiyi seçme hedefi üzerinde durulmaktadır. Bu çalışma, İzmir'deki 13 Organize Sanayi Bölgesi (OSB)'nde yer alan 47 farklı şirket için itme-çekme stratejilerine yönelik belirlenen kriterlerin önceliklendirilmesine odaklanmıştır. Kriterler, genel kriterler, itme kriterleri ve çekme kriterleri olmak üzere üç kategoride değerlendirilmiştir. OSB'nin üretim faktörlerine göre maksimum faydayı sağlayan kriterlerin değerlendirilmesi ve bu kriterler arasındaki ilişkilerin analizi için CRITIC (Criterion-Induced Weighting for Technique for Order Preference by Similarity to Ideal Solution) ve DEMATEL (Decision Making Trial and Evaluation Laboratory yöntemlerini içeren entegre bir yaklaşım önerilmiştir. DEMATEL yöntemi, anket yanıtlarına dayalı ikili karşılaştırmalar gerçekleştirmek için kullanılırken, CRITIC yöntemi anket yanıtlarının çelişki analizini yapmak için kullanılmıştır. Bu çalışmanın amacı, DEMATEL yöntemi ile elde edilen ağırlık değerlerini CRITIC çelişki analizi ile dengelemektir. Ayrıca, DEMATEL yöntemiyle kriter ağırlıklarının hesaplanması için yeni bir yaklaşım önerilmiştir. Bu iki yöntemden elde edilen ağırlık değerleri ile hibrit ağırlık değerleri elde edilmiştir. Entegre yöntemden elde edilen sonuçlara göre, en önemli kriterler; genel kriterler arasında en yüksek ağırlığa sahip olan "Doğru Miktarlarda Teslimat", itme kriterleri arasında en yüksek ağırlığa sahip olan "Süreçte Ürün Yeterlilik Durumu" ve çekme kriterleri arasında en yüksek ağırlığa sahip olan "Üretimde Aksaklık" olarak belirlenmiştir. Bu çalışma, İzmir'deki şirketler için itme-çekme stratejilerinin belirlenmesine yönelik önceliklendirilmiş kriterler sunarak, lojistik süreçlerin iyileştirilmesine ve şirketlerin rekabet gücünün artırılmasına katkı sağlamayı amaçlamaktadır. Bu bağlamda, çalışma şirketlerin operasyonel maliyetlerini azaltma, üretim süreçlerini optimize etme ve tedarik zinciri performansını artırma gibi hedeflere ulaşmalarına yardımcı olabilir.

Noise pollution caused by transportation has become problem for societies with increasing traffic volumes and the development of cities day by day. In European countries and Turkey, transportation-related noise pollution has caused health problems as well as reducing the living standards of the people. It is necessary to carry out detailed studies in order to determine the problems caused by noise pollution, especially in the regions where industry and labor force are concentrated, and to solve these problems. The city of Izmir is one of the cities where the noise caused by the railway needs to be investigated. As a result of the studies carried out in the research area, the railway noise pollution map was created with the IDW method with the data obtained from 8 different points. Afterwards, the effect of railway systems on the land usage types of cities, noise pollution problems, and the period between 2000-2018 were examined. The negative effects of noise pollution caused by the railway in the hinterland of the suburban line in the study area and its effect on the land usage type were investigated with the EU and Turkey Noise Regulation and the land cover data of the European Environment Agency.

The fact that transportation infrastructure has a great impact on economic development reveals the necessity of more economical, comfortable, and environmentally friendly transportation between transportation systems. With the rapid development of technology, the importance of the railway transportation system, which can carry high-capacity, economical and safe cargo, in the transportation sector is increasing. Although railway transportation has been used for many years in our country, it is not at the desired level in terms of both freight and passenger transportation. For a more balanced and systematic distribution of transportation infrastructure, transportation systems should be analyzed. In this study, the development of railway transport in Turkey, its ratio among other types in terms of freight and passenger transport, and its status in railway transport compared to European countries have been analyzed. In line with this information, the deficiencies in rail transport and the investments that need to be made have been determined.