Scipedia
Not a member yet
33380 research outputs found
Sort by
Classical and Bayesian Stress-Strength Reliability Estimation for Weibull Data under Unified Hybrid Censoring Scheme with LED Application
No full textThe stress-strength reliability parameter is a key metric used in various fields, including engineering, medicine, and business. In engineering, it quantifies the probability that a system’s strength X exceeds the applied stress Y . In this study, we examine for the first time four estimation approaches for evaluating the stress-strength reliability parameter R= P(Y < X ), where X and Y are independent Weibull random variables with different scale parameters but a common shape parameter. The analysis is conducted under a unified hybrid censoring scheme. From the classical perspective, we employ the maximum likelihood and maximum product of spacings methods to obtain both point and interval estimates. From the Bayesian perspective, two forms of the posterior distribution, based on the likelihood and spacings functions, are derived and analyzed using Markov Chain Monte Carlo sampling techniques. The Bayes estimates of R are obtained under the symmetric squared error loss, and the corresponding Bayesian credible intervals are also computed. To compare the four point estimators and the four interval estimators, an extensive simulation study is performed using various experimental scenarios. Finally, comprehensive analyses for organic white light-emitting diode datasets mixed with three colors, namely red, green, and blue, are provided.OPEN ACCESS Received: 29/06/2025 Accepted: 16/09/2025 Published: 23/01/202
Statistical Inference of Step Stress Partially Accelerated Life-Testing for Insulating Fluid between Electrodes under Censored Data and Different Loss Functions
No full textLong testing times are usually required for the life testing of very reliable products or materials. The testing process can be hastened by using accelerated life tests. The lifespan of the items that accelerated life tests inspect is reduced since they test products in more severe circumstances than those found in regular use scenarios. Data that was censored and disclosed the precise timings of failure may point to accelerated life tests where all units assigned to test are unknown, or where all units assigned to test have not failed for a few reasons, including challenges with technology, tools, costs, and schedules. The step-stress partially accelerated life test was examined in this work using the type-I progressive hybrid censoring scheme and the type-II progressive censoring scheme. The influence of the stress shift is explained using the tempered random variable model, where the failure times of the items are assumed to follow the alpha power Lomax distribution. The unknown parameters are estimated using the maximum likelihood estimation and Bayesian methods. The asymptotic theory of maximum likelihood estimation is also employed in the construction of the approximate confidence intervals. While the point estimates under two censoring schemes are compared in terms of absolute biases and root mean squared errors, approximate confidence intervals and coverage probabilities are compared in terms of their lengths and coverage probabilities. Additionally, three possible optimal test strategies are investigated using different optimal criteria. The performance of the estimators was evaluated and contrasted with two censoring techniques with various sample sizes using a simulation study. Finally, a numerical example for insulating fluid between electrodes data is presented to illustrate how the methods will work in real-world scenarios.OPEN ACCESS Received: 11/06/2025 Accepted: 29/07/2025 Published: 23/01/202
Automatised Optimisation of Material Joints for Additive Manufactured Multimaterial Components
No full text
Heat Transfer and Electroosmotic Flow over Stretching Sheet: A Sensitive Analysis through Response Surface Method
No full textRecent advancements in electro-osmotic surface coatings have led to significant theoretical and numerical exploration of how zeta potential influences the electroosmotic flow of viscous ionic fluids over a stretching sheet. The governing boundary layer equations are derived from the fundamental laws of mass, momentum, and energy conservation using appropriate similarity transformations and non-dimensionalization techniques. This system of equations is solved numerically using MATLAB’s bvp4c solver. The accuracy of the computational results is confirmed through comparison with previously published studies. To better understand the influence of various parameters on flow and thermal behavior, Response Surface Methodology and Factorial Plot analysis are applied. These statistical tools enable sensitivity analysis by systematically investigating the effects of zeta potential, electroosmosis parameter, electric field strength, and Prandtl number on key flow characteristics such as velocity, temperature distribution, skin friction coefficient, and Nusselt number. The results reveal that the electric field parameter plays a dominant role in enhancing axial velocity and increasing skin friction, making it a key factor in flow dynamics. The zeta potential significantly influences the boundary layer by modifying the electrical double layer and surface charge distribution, leading to noticeable deceleration. Meanwhile, the Prandtl number primarily governs thermal gradients and heat transfer rates, controlling the thermal behavior of the fluid. These physical insights, combined with the optimization capability of Response Surface Methodology, provide actionable guidelines for the design of electroosmotic coating processes and lab-on-chip biomedical devices.OPEN ACCESS Received: 07/08/2025 Accepted: 17/10/2025 Published: 03/02/202
AI-Driven Multimodal Analysis of User Experience in Immersive Environments: A Case Study of The Sphere
No full textArtificial intelligence (AI) and sensing technologies are reshaping how people experience immersive environments. This study investigates how audiences perceive and emotionally respond to such environments through an AI-driven mixed-methods analysis. A dataset of 275 usergenerated YouTube videos documenting experiences with The Sphere, an AI-convergent immersive environment, totaling over 3000 min of content and 24 million cumulative views, was analyzed to extract experiential themes, dominant emotions, and their relationships with public engagement metrics. The analysis identified seven key experiential themes: Awe of the Display, Personalized Spatial Audio Experience, Full-Body Sensory Engagement, Dynamic Visual Spectacle, Joyful Human–AI Encounter, Futuristic Spatial Design Experience, and Transformative Event Environment. Sentiment analysis revealed that fear was the most dominant emotion in textual narratives (42.3%), followed by surprise, sadness, happiness, and anger, whereas video-based analysis highlighted happiness (25.8%) and sadness (24.5%) as the most salient visual emotions. This contrast suggests that linguistic expressions emphasized feelings of awe and overwhelm, while visual cues reflected affective immersion and emotional depth. Regression results showed that Awe of the Display had the strongest positive impact on engagement (views, likes, comments), while Personalized Spatial Audio Experience showed a negative effect. These findings deepen the understanding of user experience in immersive environments and demonstrate how AI-assisted multimodal analysis can reveal the dynamics between audience perception and engagement in next-generation immersive environments.OPEN ACCESS Received: 12/11/2025 Accepted: 15/12/2025 Published: 03/02/202
AI-Driven Predictive Analytics for Demand Forecasting in Transportation Logistics to Enhance Supply Chain Agility
No full textThe logistics and transportation sectors are struggling with major issues like demand variations, disruptions, and inefficiencies, which ultimately undermine the agility and efficiency of the entire supply chain. Most of the time, traditional forecasting models are not entirely accurate in response to life-changing factors like weather, traffic, and inventory levels. The present research intends to build an AI-powered predictive model that can seamlessly enhance not only demand forecasting and logistics but also by the integration of real-time data. The framework incorporates several Machine Learning (ML) models, which are Light GBM for demand forecasting, Random Forest for disruption prediction, Linear Regression for shipping cost estimation, and Support Vector Regression for delivery time deviation prediction. A thorough dataset containing historical demand, weather conditions, traffic, and stock levels was used for the model’s training and evaluation, and its performance was monitored using MAE, MSE, RMSE, and MAPE metrics. The findings indicate that the suggested framework is a lot better than the existing ones, with Light GBM getting the lowest MAE (0.056), MSE (0.005), RMSE (0.072), and MAPE (0.142). This means that the new system can predict much better than before, thus making it possible for the company to take the right decision at the right time and consequently improving the overall supply chain efficiency. The research paper reveals the future possibilities of AI-based solutions for optimising logistics operations and building supply chain resilience.OPEN ACCESS Received: 20/10/2025 Accepted: 25/12/2025 Published: 03/02/202
