5916 research outputs found
Sort by
Performance characterization of vehicular visible light communication in the presence of sunlight
No full textVehicular visible light communication (VLC) has emerged as a promising alternative to radio-based vehicular communication and aims to utilize the widely available light emitting diode (LED) based vehicle exterior lighting for dual-purpose illumination and data transmission. While VLC has thrived in indoor environments and has already been commercialized, its adoption in outdoor settings faces unique challenges. In particular, solar irradiance significantly impacts the performance of vehicular VLC systems and exhibits its effect by increasing background noise and elevating the direct current (DC) level in the photodetector output. In this paper, we conduct a comprehensive performance characterization of vehicular VLC in the presence of sunlight. We utilize the empirical ASHRAE clear-sky solar irradiation model to estimate background noise, incorporating geographical, temporal, and vehicular mobility factors to account for changing sunlight conditions. We derive a closed-form expression for signal-to-noise ratio (SNR) as a function of solar irradiance, receiver parameters (e.g., aperture diameter, responsivity, and bandwidth), and vehicular positions relative to each other. We conduct an experimental study in outdoor vehicular environments to validate the theoretical models, demonstrating their practical applicability under varying sunlight conditions.Tamkeen under the Research Institute NYUAD ; EU Horizon 2020 MSCA-ITN (ENLIGHT'EM)Publisher versio
A matheuristic for the joint replenishment problem with and without resource constraints
No full textWe study the Joint Replenishment Problem (JRP), which arises from the need for coordinating the replenishment of multiple items that share a common fixed cost. Even in the basic setting, determining the optimal replenishment plan is an NP-Hard problem. We analyze both the JRP under indirect grouping policy and its variant with restrictions like transportation capacity, budget capacity, and item transportation compatibility. Additionally, we consider uncertainty characteristics such as imperfect item quality, as highlighted in related literature studies. We propose a novel matheuristic method that determines the best basic cycle time while addressing the problem with a fixed cycle time using a linear integer model. The proposed method is quite versatile to handle additional real-life constraints effectively. Based on an extensive computational study, we conclude that for the basic setting under indirect grouping policy, the proposed algorithm outperforms the benchmark algorithms in the literature by 0.3% on average. For more complicated settings with additional restrictions, our proposed algorithm outperforms the benchmark algorithm by around 5% on average
Two-echelon prize-collecting vehicle routing with time windows and vehicle synchronization: A branch-and-price approach
No full textThe steady growth in e-commerce and grocery deliveries within cities strains the available infrastructure in urban areas by increasing freight movements, aggravating traffic congestion, and air and noise pollution. This research introduces the Two-Echelon Prize-Collecting Vehicle Routing Problem with Time Windows and Vehicle Synchronization, where deliveries are carried out by smaller low- or zero-emission vehicles and larger trucks. Given their capacity restrictions, the smaller vehicles can only deliver small-sized orders and must be replenished via depot locations or larger-sized trucks. Besides replenishing smaller vehicles at satellite locations, larger trucks can deliver small orders and larger items. Managing these two types of fleets in an urban setting under consideration of capacity limitations, tight delivery time windows, vehicle synchronization, and selective order fulfillment is challenging. We model this problem on a time-expanded network and apply network reduction by considering the time window constraints. In addition, we propose a branch-and-price algorithm capable of solving instances with up to 200 customers, which continuously outperforms a state-of-the-art general-purpose optimization solver. Moreover, we present several managerial insights concerning synchronization, vehicles, and the placement of depot/satellite locations.Publisher versio
A graphlet-based explanation generator for graph neural networks over biological datasets
No full textBackground Graph neural networks' (GNNs) explainability, especially the explanation of edges and interactions among vertices in GNNs, is demanding mainly owing to dynamics and groupings between vertices. The existing graph explainability methods ignore the analysis of the following tasks weights over subgraphs but instead analyze solely sample-level explainability. Such sample-level explainability decreases their generalizability since it directly searches the explaining behaviour in the input dataset.Objective In this study, we come up with a novel Orbit-based GNN explainer (OExplainer), which integrates both sample-level and method-level approaches over a predetermined set of subgraphs. As part of such analysis of subgraphs, our goal is to interpret graphs more comprehensively and intelligibly while providing each vertex's explainability score for a particular graph instance.Methods Our OExplainer decomposes the following graph neural network weights into explaining subgraph bases while identifying and characterizing particular predictions. By such characterization, we can carefully and accurately interpret the predetermined graph orbit's role in vertex representation determination. In this characterization, we can also clarify the method's behaviour generally for the whole input dataset. Moreover, we come up with novel vertex-specific scores in our subgraph-based approach over nonisomorphic graphlets. Such vertex-specific score encourages sample-level vertex improvement, and such improvement is related to the graph neural network's vertex classification task.Results Our experiments over simulated datasets confirm the importance and criticality of method weights in vertex classification explanation. In this case, method weight decomposition also has criticality. Our detailed experiments over multiple real protein-protein interaction datasets and metabolic interaction networks also exhibit enhanced performance in vertex classification.Conclusion In both simulated and biological protein-protein interaction datasets, our approach outperforms the competing explanation approaches.TÜBİTA
Numerical simulations of non-fluorescent states of carboxyfluoresceins in presence of heavy iodine ions
No full textIn the present research, the effects of heavy iodine ions on dark transient state populations of carboxyfluorescein derivatives, each with a different number of bromine atoms attached, were computationally examined via numerical simula tions adapted to a widefield fluorescence microscopy integrated with a microfluidics platform. Numerical simulations were car ried out by considering geometrical profile of excitation laser beam, microscopy and microfluidics parameters of a proposed experimental design as well as electronic transition rates of fluorescent molecules. Electronic state model of studied dyes was treated as a system of first order ordinary differential equa tions and time-dependent solutions of long-lived, non-fluorescent dark state populations were computed for each dye at varying potassium-iodine [KI] concentrations. Analytical solutions of state populations were then adapted to a proposed experimental setup to systematically analyze how dark state populations evolve when carboxyfluorescein derivatives pass through exci tation beam field in a microfluidics chip under different flow speeds. Computational experiments have successfully uncovered systematic changes in dark triplet and photo-oxidized state populations upon the addition of iodine ions into fluorophore solutions.TÜBİTAKPost prin
Dynamic and precise electromagnetic levitation of single cells
No full textThe biophysical properties of single cells are crucial for understanding cellular function and behavior in biology and medicine. However, precise manipulation of cells in 3-D microfluidic environments remains challenging, particularly for heterogeneous populations. Here, we present "Electro-LEV," a unique platform integrating electromagnetic and magnetic levitation principles for dynamic 3-D control of cell position during separation. We demonstrated that small current adjustments in electromagnets significantly alter the levitation heights of diverse particles and cell types. By periodically modulating and tracking cell positions along the z-axis, Electro-LEV identified distinct levitation behaviors between single cells and cell clusters, with clusters responding more rapidly to magnetic field changes. Furthermore, we demonstrated that Electro-LEV significantly enhances the purity and efficiency of levitational sorting, achieving 10-fold enrichment of live cells from 50% starting viability samples and 18.8-fold enrichment from 10% starting viability samples. These results establish Electro-LEV as a powerful tool for investigating cellular heterogeneity, differentiating cell sizes and types, and improving cell sorting efficiency. Thus, Electro-LEV is broadly applicable, offering different possibilities for high-resolution cell analysis and label-free cell sorting in various biomedical fields, including but not limited to single-cell sequencing and drug screening.Career Award at the Scientific Interface ; Burroughs Wellcome Fund ; Gordon & Betty Moore Foundation ; McCormick and Gabilan Faculty Award from Stanford University ; Donald E. and Delia B. Baxter Foundatio
How far back does medical consent go? A journey through ottoman legal records in Ottoman Empire
No full textTracing the evolution of informed consent from the Hippocratic tradition to the Ottoman Empire reveals its enduring role as a fundamental ethical principle supporting patient autonomy. Spanning diverse medical and cultural landscapes-including Ancient Greece, Byzantium, Islamic medicine and Ottoman legal practices-this historical trajectory uncovers a continuous and evolving dialogue between physicians and patients. It reflects a persistent recognition of the moral and practical necessity for physicians to share medical information and for patients to engage voluntarily in decisions regarding their health. A particularly significant historical juncture is found in the Ottoman Empire. Especially noteworthy are 16th-century and 17th-century court records from Istanbul, which provide some of the earliest concrete examples of formalised patient consent in the Ottoman era. These records, found in qadi (kad & imath;) sicilleri (Ottoman court registers), document patients giving explicit permission-often witnessed and recorded-for surgical procedures. In our study, we employed a methodology that began with a systematic review of relevant national and international literature, followed by the examination of qadi court records from the years 1579 to 1663. Through a concept-based search, 21 documents from the Istanbul Province's kad & imath; registers were located, analysed and evaluated in terms of their informed consent elements. Such documentation demonstrates that informed consent, although not institutionalised in the modern sense, was already being practised as a legally and ethically meaningful process. These examples represent not only a cultural and legal continuity from earlier historical practices but also a key moment in the formal recognition of the patient's voice in medical decision-making. By analysing these sources, the study not only reveals how these legal responsibilities were understood and practised in premodern contexts, but also offers valuable historical insight that enriches contemporary discourse on medical ethics and law. This continuity shows that informed consent is not merely a modern legal formality, but a long-standing ethical commitment embedded across cultures and civilisations
Utilization of district morphology for optimal energy performance: A case study in ankara, Turkey
No full textThis research investigates the influence of district morphology on solar radiation, energy performance, and renewable energy potential. The case district area in Ankara, Turkey, has an area of 76,100 m(2) and 15 apartments and includes almost 2,400 dwellers. The methodology comprises three phases: computational modeling, performance simulations, and comparative assessments. Twenty-eight alternative district models were developed and analyzed using Rhinoceros 7.0 with the combination of Grasshopper plug-ins-Ladybug 1.8.0, Honeybee, and Climate Studio 1.9. The findings highlight that the district scenario with an irregular Y-axis configuration performs most satisfactorily regarding solar radiation, building energy, and photovoltaic panel integration. This alternative model generates the highest electricity from solar canopies, approximately 3.5 times more than PVs on roofs. The model results in a significant reduction of approximately 35% in the district's total primary energy consumption compared to the base scenario
An integrated matlab code for homogenization-based topology optimization and generating functionally graded surface lattices for additive manufacturing
No full textTriply periodic minimal surfaces (TPMS) lattices are gaining popularity for enhancing structural efficiency in many engineering applications. Functionally graded TPMS structures provide more customized mechanical properties and improved functionality compared to typical homogenous designs by deliberately altering material properties throughout the lattice. This study presents a novel framework by integrating a homogenization-based topology optimization method with functionally graded lattice creation, utilizing a streamlined and versatile MATLAB code. The methodology encompasses several essential phases, including preprocessing, finite element analysis, sensitivity analysis, density filtering, optimization, element density visualization, and lattice reconstruction. These steps facilitate the development of highly efficient lattice structures with varied attributes, rendering them optimal for additive manufacturing and full-scale analysis. To ensure the accuracy of the established methodology, three optimization case studies with different boundary conditions are defined, and the mechanical reactions of the optimized lattice structures in filled with different TPMS structures are extensively validated by comparing them to both full-scale finite element models and experiments. The comparative results demonstrate that the mechanical responses obtained from topological analysis closely correspond to those acquired from full-scale models and experiments.TÜBİTAKPublisher versio