Polygenic risk score (PRS) tools differ substantially in statistical assumptions, input requirements, and implementation complexity, making direct comparison difficult. We developed a harmonized, implementation-aware benchmarking framework to evaluate 46 PRS tools across seven binary UK Biobank phenotypes and one continuous trait under three model configurations: null, PRS-only, and PRS plus covariates. The framework integrates standardized preprocessing, tool-specific execution, hyperparameter exploration, and unified downstream evaluation using five-fold cross-validation on high-performance computing infrastructure. In addition to predictive performance, we assessed runtime, memory use, input dependencies, and failure modes. A Friedman test across 40 phenotype--fold combinations confirmed significant differences in tool rankings ($χ^2 = 102.29$, $p = 2.57 \times 10^{-11}$), with no single method universally optimal. These findings provide a reproducible framework for comparative PRS evaluation and demonstrate that tool performance is shaped not only by statistical methodology but also by phenotype architecture, preprocessing choices, covariate structure, computational demands, software robustness, and practical implementation constraints.
Eva Nursifa Fauziah Suwandi, Bambang Suhardi, Etika Muslimah
This study examines human factors in bus accidents using bibliometric analysis to identify publications, research trends, and collaborative relationships between authors and institutions. Data from 1.834 publications in the Scopus database during the period 2014–2024 were analyzed using VOS viewer software to display network visualization and density. The results show a significant increase in publications since 2018, with a peak in 2022. Research related to human factors in bus accidents is growing, with a focus on fatigue, distraction, and driver behavior. The keyword and density analysis identified that “Human” was the most frequently discussed topic. China is the country with the highest contribution, and the medical field plays a major role in this topic. These findings highlight the importance of understanding human factors in efforts to improve transportation safety.
Arfa Ijaz, Saleha Qureshi, Ubaid Ur Rehman Zia
et al.
The global steel industry emits 1.92 tons of CO<sub>2</sub> per ton of output and faces urgent pressure to decarbonize. In Pakistan, the sector accounts for 0.29 tons of CO<sub>2</sub> per ton of output, with limited mitigation frameworks in place. Green hydrogen (GH2)-based steelmaking offers a strategic pathway toward decarbonization. However, realizing its potential depends on access to renewable energy. Despite Pakistan’s substantial technical wind potential of 340 GW, grid limitations currently restrict wind power to only 4% of national electricity generation. This study explores GH<sub>2</sub> production through sector coupling and power wheeling, repurposing curtailed wind energy from Sindh to supply Karachi’s steel industry, and proposing a phased roadmap for GH, enabling fossil fuel substitution, industrial resilience, and alignment with global carbon-border regulations.
Giacomo Cangi, Alessandro Angeletti, Massimiliano Palmieri
et al.
Understanding and knowing the extent of dynamic loads (forces and moments) acting on critical points of structures from the early stages of new product design allows companies to reduce product time-to-market and cut costs associated with physical validation tests. This is made possible through the use of commercial software that enables the simulation of the dynamic behavior of a wide range of mechanical systems, and beyond. However, the use of such software is not straightforward and often requires a specialist within the company to have in-depth knowledge of both simulations and the software itself. Consequently, many companies give up the use of these tools, compromising the whole product development process, falling into the design and testing iterative loop. The multibody mathematical model developed in this study, simulated using python programming language, allows easy retrieval of all the necessary information without requiring the user to be a specialist in multibody dynamics simulations. This model is intended as a tool specifically designed for the type of product (aerial work platform) that serves, during the conceptual and preliminary design phases, to predict dynamic loads through calculations. The balance between the simplicity of such a tool and the accuracy of the results is the key point for the success of this work.
A driving force for the realization of a sustainable energy supply is the integration of renewable energy resources. Due to their stochastic generation behaviour, energy utilities are confronted with a more complex operation of the underlying power grids. Additionally, due to technology developments, controllable loads, integration with other energy sources, changing regulatory rules, and the market liberalization, the systems operation needs adaptation. Proper operational concepts and intelligent automation provide the basis to turn the existing power system into an intelligent entity, a cyber-physical energy system. The electric energy system is therefore moving from a single system to a system of systems. While reaping the benefits with new intelligent behaviors, it is expected that system-level developments, architectural concepts, advanced automation and control as well as the validation and testing will play a significantly larger role in realizing future solutions and technologies. The implementation and deployment of these complex systems of systems are associated with increasing engineering complexity resulting also in increased engineering costs. Proper engineering and validation approaches, concepts, and tools are partly missing until now. Therefore, this paper discusses and summarizes the main needs and requirements as well as the status quo in research and development related to the engineering and validation of cyber-physical energy systems. Also research trends and necessary future activities are outlined.
This study assesses the capability of ChatGPT to generate finite element code for geotechnical engineering applications from a set of prompts. We tested three different initial boundary value problems using a hydro-mechanically coupled formulation for unsaturated soils, including the dissipation of excess pore water pressure through fluid mass diffusion in one-dimensional space, time-dependent differential settlement of a strip footing, and gravity-driven seepage. For each case, initial prompting involved providing ChatGPT with necessary information for finite element implementation, such as balance and constitutive equations, problem geometry, initial and boundary conditions, material properties, and spatiotemporal discretization and solution strategies. Any errors and unexpected results were further addressed through prompt augmentation processes until the ChatGPT-generated finite element code passed the verification/validation test. Our results demonstrate that ChatGPT required minimal code revisions when using the FEniCS finite element library, owing to its high-level interfaces that enable efficient programming. In contrast, the MATLAB code generated by ChatGPT necessitated extensive prompt augmentations and/or direct human intervention, as it involves a significant amount of low-level programming required for finite element analysis, such as constructing shape functions or assembling global matrices. Given that prompt engineering for this task requires an understanding of the mathematical formulation and numerical techniques, this study suggests that while a large language model may not yet replace human programmers, it can greatly assist in the implementation of numerical models.
Large Language Models (LLMs) are being increasingly used in software engineering tasks, with an increased focus on bug report resolution over the past year. However, most proposed systems fail to properly handle uncertain or incorrect inputs and outputs. Existing LLM-based tools and coding agents respond to every issue and generate a patch for every case, even when the input is vague or their own output is incorrect. There are no mechanisms in place to abstain when confidence is low. This leads to unreliable behaviour, such as hallucinated code changes or responses based on vague issue reports. We introduce BouncerBench, a benchmark that evaluates whether LLM-based software agents can refuse to act when inputs are ill-defined or refuse to respond when their own outputs are likely to be incorrect. Unlike prior benchmarks that implicitly incentivize models to generate responses even when uncertain, BouncerBench aims to improve precision by targeting two overlooked failure points: (1) vague or underspecified issue descriptions in tickets and (2) logically or functionally incorrect code patches created by the system. It measures whether proposed systems can distinguish actionable issues from vague tickets and valid patches from untrustworthy ones. We also implement a basic input and output bouncer, evaluating how well current LLMs can abstain when needed. Our results show that most models fail to abstain from underspecified inputs or incorrect outputs. Hence, we conclude that there is significant room for improvement before LLMs can be trusted to make correct decisions and recommendations in real-world software engineering workflows. BouncerBench provides a first step toward evaluating and building more cautious, trustworthy code agents. The replication package, dataset, and leaderboard can be found at bouncerbench.com
Ivelina Balabanova, Kristina Sidorova, Georgi Georgiev
In the paper, personalized results are presented in the methodology for monitoring information security based on voice authentication. Integration of sound preprocessing and Machine Learning techniques for feature extraction, training, and validation of classification models has been implemented. The objects of research are staked mixed-test voice profiles. Classifies were selected with quantitative evaluation under a threshold of 90.00% by Naive Bayes and Discriminant Analysis. Significantly improved accuracy to approximate levels of 96.0% was established at Decision Tree synthesis. Strongly satisfactory performance indices were reached at the diagnosis of voice profiles using Feed-Forward and Probabilistic Neural Networks, respectively, 98.00% and 100.00%.
With the development of the Internet of Things (IoT) and communication networks, the concept of a smart city emerged spontaneously. In the traffic control of a smart city, the vehicular network plays an important role. If the driving information of vehicles can be collected from the vehicular network and then aggregated into a smart city system, traffic control facilities can be adjusted in real time to improve traffic or increase traffic safety. This study is based on WAVE/DSRC under the IEEE 802.11p and IEEE 1609 standards. When the vehicle is moving, the On-Board Unit (OBU) on the vehicle and the Roadside Unit (RSU) transmit data through the 5.9 GHz (5.85–5.925 GHZ) frequency band to establish vehicle-to-vehicle communication (V2V) and vehicle-to-infrastructure communication (V2I). The immediacy of the transmission of beacon messages is also discussed over a vehicular network. A beacon message is essential for communication in a vehicular network, including various safety messages, such as driving directions, driving speed, and location. We introduce the Age-of-Information (AoI) indicator to reflect the immediacy of information. AoI is used for the elapsed time after the sender samples the message until the receiver receives the message. We propose a centralized AoI-based protocol and a decentralized AoI-based protocol. By using Random-Walk and SUMO, we simulate driving dynamics and various RSU setting scenarios. Finally, we verify the performance of the proposed AoI-based protocol through experimental simulations.
A multibladed drag-based Vertical Axis Wind Turbine (VAWT) was developed and its startup dynamics evaluated using wind tunnel tests. The experimental data obtained for the time-based angular position of the rotor shaft at Aberdeen’s average wind speed of 6 m/s show an initial rapid acceleration of the VAWT due to the drag force being exerted on the rotor blades. This acceleration becomes more gradual until the VAWT reaches its peak rotational speed of 85 rpm in 30 s, which corresponds to an operating tip speed ratio (TSR) of 0.42. The operating TSR of the VAWT was found to be 27% higher than previously reported in numerical studies.
Instrumented wheelsets are widely used in the railway industry for the purpose of measurement of the wheel-rail interaction forces, which are crucial factors in a running safety assessment. The lateral force, which is the lateral component of the wheel-rail interaction forces, is measured using strains induced by bending deformation of the wheel web in typical configuraions of instrumented wheelset. In addition to the use of bending deformation, the author has proposed a new configuration of instrumented wheelset which usitilizes shear strains of wheel web as a measure of lateral force. In these principles of lateral force measurement using bending or shear strains, the measurement accuracy descreases when a loading point of wheel load, which is the vertical component of wheel-rail interaction forces, is shifted laterally since the wheel load also induces the bending and shear strains. In this research, output characteristics of several types of bridge circuits, which are categolized as “axially-asymmetric bridge circuit,” are investigated for the purpose of reducing the influence of wheel load and improving the measurement accuracy of lateral force. Numerical investigations using a finite element method and experimental investigations are carried out and the results show that the axially-asymmetrization of the bridge circuits is effective to reduce the infuluence of wheel load.
Mechanical engineering and machinery, Engineering machinery, tools, and implements
Many optimization problems in engineering and industrial design applications can be formulated as optimization problems with highly nonlinear objectives, subject to multiple complex constraints. Solving such optimization problems requires sophisticated algorithms and optimization techniques. A major trend in recent years is the use of nature-inspired metaheustic algorithms (NIMA). Despite the popularity of nature-inspired metaheuristic algorithms, there are still some challenging issues and open problems to be resolved. Two main issues related to current NIMAs are: there are over 540 algorithms in the literature, and there is no unified framework to understand the search mechanisms of different algorithms. Therefore, this paper attempts to analyse some similarities and differences among different algorithms and then presents a generalized evolutionary metaheuristic (GEM) in an attempt to unify some of the existing algorithms. After a brief discussion of some insights into nature-inspired algorithms and some open problems, we propose a generalized evolutionary metaheuristic algorithm to unify more than 20 different algorithms so as to understand their main steps and search mechanisms. We then test the unified GEM using 15 test benchmarks to validate its performance. Finally, further research topics are briefly discussed.
The effectiveness of model-driven software engineering (MDSE) has been successfully demonstrated in the context of complex software; however, it has not been widely adopted due to the requisite efforts associated with model development and maintenance, as well as the specific modelling competencies required for MDSE. Concurrently, artificial intelligence (AI) methods, particularly deep learning methods, have demonstrated considerable abilities when applied to the huge code bases accessible on open-source coding platforms. The so-called big code provides the basis for significant advances in empirical software engineering, as well as in the automation of coding processes and improvements in software quality with the use of AI. The objective of this paper is to facilitate a synthesis between these two significant domains of software engineering (SE), namely models and AI in SE. The paper provides an overview of the current state of AI-augmented software engineering and develops a corresponding taxonomy, ai4se. In light of the aforementioned considerations, a vision of AI-assisted big models in SE is put forth, with the aim of capitalising on the advantages inherent to both approaches in the context of software development. Finally, the pair modelling paradigm is proposed for adoption by the MDSE industry.
This paper presents HyperGraphOS, a significant innovation in the domain of operating systems, specifically designed to address the needs of scientific and engineering domains. This platform aims to combine model-based engineering, graph modeling, data containers, and documents, along with tools for handling computational elements. HyperGraphOS functions as an Operating System offering to users an infinite workspace for creating and managing complex models represented as graphs with customizable semantics. By leveraging a web-based architecture, it requires only a modern web browser for access, allowing organization of knowledge, documents, and content into models represented in a network of workspaces. Elements of the workspace are defined in terms of domain-specific languages (DSLs). These DSLs are pivotal for navigating workspaces, generating code, triggering AI components, and organizing information and processes. The models' dual nature as both visual drawings and data structures allows dynamic modifications and inspections both interactively as well as programaticaly. We evaluated HyperGraphOS's efficiency and applicability across a large set of diverse domains, including the design and development of a virtual Avatar dialog system, a robotic task planner based on large language models (LLMs), a new meta-model for feature-based code development and many others. Our findings show that HyperGraphOS offers substantial benefits in the interaction with a computer as information system, as platoform for experiments and data analysis, as streamlined engineering processes, demonstrating enhanced flexibility in managing data, computation and documents, showing an innovative approaches to persistent desktop environments.
Yallappa Dengeru, Kavitha Ramasamy, Surendrakumar Allimuthu
et al.
Insecticide applications are typically being carried out with traditional manual spraying equipment in redgram, which leads to inadequate control of insects due to higher crop height. The modern deployment of tractor-drawn spray machines causes serious damage to the crop. In this connection, unmanned aerial vehicle (UAV) spray technology has great potential for precise insecticide application in redgram crops. One of the important machine parameters influencing droplet deposition and drift characteristics in UAV sprayers is downwash airflow generated by a multi-rotor propeller. A field experiment was carried out at the redgram research field (N11.01, E76.92), Tamil Nadu Agricultural University, Coimbatore, Tamil Nadu, during 2021–2022 to study the spray drift and deposition characteristics of an autonomous UAV sprayer. The Imidacloprid (a.i. 17.8SL) insecticide mixed with water in a ratio of 1 mL per liter was sprayed with a UAV sprayer. Water-sensitive paper samples were kept at upper, middle, and bottom positions on the leaves, and data were analyzed for the spray droplet size, deposition rate, droplet density, and area coverage both in target and non-target areas using Spray Deposit Scanner software. UAV spray droplet deposition rate (2.93 ± 0.17, 2.01 ± 0.08, and 2.21 ± 0.162.38 μL cm−2), droplet density (47 ± 4.04, 53 ± 3.61, and 52 ± 8.74 droplets cm−2), and area coverage (15.72 ± 0.39, 16.60 ± 0.71, and 14.99 ± 0.39%) were highest in the upper layer as compared to the middle layer (droplet deposition rate: 1.21 ± 0.08, 1.07 ± 0.03, and 0.77 ± 0.02 μL cm−2; droplet density: 42 ± 2.52, 43 ± 8.50, and 38 ± 2.52 droplets cm−2; area coverage: 10.95 ± 0.81, 11.22 ± 0.56, and 8.57 ± 0.44%) and bottom layer (droplet deposition rate: 0.41 ± 0.06, 0.35 ± 0.03, and 0.33 ± 0.03 μL cm−2; droplet density: 22 ± 4.36, 17 ± 3.51, and 19 ± 4.51 droplets cm−2; area coverage: 2.78 ± 0.29, 2.95 ± 0.45, and 2.46 ± 0.20%, respectively). In the spray drift test, there was a higher droplet deposition rate (1.63 ± 0.09, 1.93 ± 0.05, and 1.82 ± 0.06 μL cm−2), area coverage (14.40 ± 0.07, 17.54 ± 0.36, and 16.42 ± 0.30%), and droplet density (46 ± 3.61, 54 ± 2.08, and 45 ± 3.21 No’s cm−2) in the target area as compared to the non-target area (droplet deposition rate: 0.88 ± 0.02, 0.46 ± 0.03, 0.22 ± 0.05, and 0.00 μL cm−2; droplet density: 23 ± 1.53, 11 ± 2.08, 6 ± 1.53, and 0.00 droplets cm−2; area coverage: 7.58 ± 0.34, 4.41 ± 0.19, 2.16 ± 0.05, and 0.00%, respectively), which may have been due to the downwash airflow produced by the multi-rotor propeller of the UAV sprayer. Finally, the UAV-based spraying technology results showed that the downwash air produced by the six-rotor propeller improved the penetrability of insecticide to crop leaves and led to a higher droplet deposition rate, droplet density, area coverage, and droplet penetrability on the upper layer, middle layer, and bottom layer of the plants.
The connected vehicles are the future of transportation. Road can be much safer when vehicles are able to communicate about the traffic and road conditions to its surrounding vehicles or if they can “talk” to each other. A smooth communication among vehicles can be helpful in terms of road safety in case of sudden speed breaks, bumps, slow moving traffic, poor visibility conditions such as fog and, importantly, wrong way traffic. Our work suggests potential applications of Vehicle-to-Vehicle (V2V) communication for improving road safety in Pakistan. We have considered one such unsignalized intersection point at main University Road, Karachi and presented an approach to make this road safer for vehicles and pedestrians using V2V technology.
Spurthi Joanna Selladurai, Neetu Srivastava, Ioannis E. Sarris
This study provides a framework to strategize localized efficient drug delivery in second-order blood flowing through porous blood vessels using machine learning algorithms. With the assumption of long blood vessels, the flow-governing equation, the Navier–Stokes equation, is reduced to a simpler model which is consistent with the lubrication theory. We solved this equation analytically with slip conditions and obtained the analytical expression of the velocity profile for the Newtonian model. We modelled the concentration of nanodrugs with an advection diffusion equation to analyze the effect of concentration on the localized disease. The particle concentration at the blood vessel wall was evaluated using the finite-difference method. To analyze the particle concentration, we implemented machine learning algorithms including Gradient Boost, XG Boost, Regression Tree, MLP Regressor, and CatBoost Regressor. Our conclusion predicts the optimum machine learning algorithm for transferring the delivery of the nanoparticle drug.
Vortex-induced vibration (VIV) is the periodic motion of a bluff body caused by fluid flow and is widely discussed in the engineering field. With the advancement of science and technology, miniaturization and integration have become the mainstream trends in biomedical chips and electronic systems, resulting in higher heat dissipation requirements per unit area. Therefore, the improvement of the heat dissipation effect of movable structures in the flow channel has been widely discussed. Among them, adding VIV motion in the microchannel generates a vortex structure, which improves heat transfer efficiency. Different from the direct displacement method of active vibration, the passive displacement of VIV is a multi-physics problem. It needs to integrate the flow field and the spring-mass system of the object for fluid–solid coupling, which greatly increases the difficulty of analysis. In this study, the Immersed-boundary method (IBM) combined with the equation of motion is used to numerically study a vortex generator that is elastically installed in a microfluidic channel and is then used to enhance the convective heat transfer of nanofluids in the channel. Unlike the common body-fitted mesh, IBM greatly reduces the computational resources required for mesh regeneration when simulating the problem of object movement in fluid–structure interaction. In addition, Buongiorno’s two-phase mixing model is used to simulate the convective heat transfer of nanofluids in microchannels by considering the Brownian motion and thermophoretic diffusion of nanoparticles in the carrier liquid. By changing the important parameters such as nanofluid concentration, Reynolds number, mass ratio, and Ur, the influence of the response characteristics of vortex-induced vibration on the heat flow field in the microfluidic channel is discussed, and the key factors for enhancing heat transfer are found out.
To study the influence of eccentricity on the dynamic characteristics of grinding electric spindle, firstly, the nonlinear oil film force (NOFF) expression of sliding bearing is established based on the boundary condition of adhesion pressure. Then, the expression of the unbalanced magnetic pull (UMP) under the eccentricity of the air gap is derived. Finally, the bearing-rotor dynamics equation of the grinding electric spindle considering the NOFF and UMP is established, and the dimensionless processing is carried out. The nonlinear dynamics study of the bearing-rotor system under the influence of geometric eccentricity and mass eccentricity is carried out through the shaft center trajectory and the shaft center vibration spectrum. The research results can provide reference for the optimal design of grinding electric spindle.
Engineering machinery, tools, and implements, Mechanical engineering and machinery