Engineering Reports

We introduce the entropy-transformed inverse Weibull distribution (ETIWL), a novel model that enhances data modeling capabilities. Through extensive simulations and real dataset applications, ETIWL demonstrates superior performance compared with benchmark distributions, offering improved reliability in survival analysis and other critical areas.

Flow chart highlighting the service mechanism of MRP with impatience and working breakdown of the server.

This study explores the integration of digitalization and lean production to enhance Overall Equipment Effectiveness (OEE) in the apparel industry. Findings indicate that targeted interventions, including process monitoring devices and digital line balancing, can increase OEE from 40% to 60% to a sustainable 75%, improving efficiency, reducing downtime, and optimizing production performance.

This work explores the unsteady radiative magnetohydrodynamic flow of a Casson fluid over an impulsively started porous plate using Caputo fractional derivatives. The model incorporates thermal radiation, magnetic field, and memory characteristics of the fluid. Results highlight how fractional parameters, magnetic strength, and radiation influence the fluid behaviors.

Approach for reduced cooling loads and kg eq-CO₂ emissions.

Based on the correspondence between top-level requirements and conceptual design parameters, the top-level requirement optimization of BWB UAVs is completed using a comprehensive evaluation method.

Enhanced analytical models for predicting long-term deflection in cracked RC beams.

The thermal performance of flat plate collectors (FPCs) was enhanced using nanofluids. A hybrid MWCNT/Fe₃O₄/water nanofluid achieved 71.9% thermal efficiency, outperforming conventional fluids. Improved Nusselt number, friction factor, and convective heat transfer highlight its effectiveness in optimizing FPC systems.

Variation of the average Nusselt number (Nu_av) with different Ra across the hot wall interface for various inner cavity shapes with wall thickness (W 0.1).

This study presents an AI-driven MPPT strategy integrating an artificial neural network and nonlinear backstepping control for grid-connected PV systems. The proposed method ensures optimal energy harvesting and seamless grid synchronization, achieving superior performance with a remarkable THD reduction to 0.2148, significantly enhancing PV system efficiency under fluctuating environmental conditions.

This study presents reliability assessments and load flow computations for the electricity grid in Rwanda at the distribution level on Gatumba and Ntongwe feeders by integrating the PV system and BESS. The power characteristics were established by utilizing the PV*SOL simulation, and quasi-dynamic and reliability simulations are conducted in DigSILENT PowerFactory.

This study presents a family of logic-minimized approximate full adders (AFA1–AFA3) designed using AND-OR gate logic and implemented on an Intel Cyclone IV FPGA, achieving up to 45.3% reduction in logic utilization, 30.3% lower power consumption, and 33.9% faster delay while maintaining high perceptual fidelity (PSNR = 34.6 dB) in image addition tasks, thereby offering an energy-efficient, FPGA-friendly arithmetic solution for error-tolerant embedded and image processing systems.

FR-FMEA combines fuzzy logic and rough set theory to address uncertainty and incomplete knowledge in failure analysis. Applied to a gas turbine system, it delivers more reliable and distinct risk rankings, identifying critical components for targeted maintenance and improved reliability.

The HPLIRF model reveals fungal disease dynamics in onion crops, identifying spore deposition and infection rates as key drivers. Optimal control combining fungicide and plant removal minimizes disease spread cost-effectively, enhancing crop resilience.

An optimized energy management system using Particle Swarm Optimization significantly improves cost-efficiency and battery stability in grid-connected PV-BESS smart grids. The proposed method outperforms linear programming and demonstrates robust performance under variable weather conditions, supporting reliable renewable integration and demand response strategies in smart energy systems.

A novel method corrects perspective distortion in street-view window extraction using a learned transformation and differentiable rendering loss, eliminating the need for vector labels and boosting accuracy.

Drilling tools are of paramount importance for mechanical metal cutting. Breakage or wear in the drilling tool can result in catastrophic failure, reduced dimensional accuracy, and poor surface finish. Therefore, it is imperative to continuously monitor drilling tools. In addition, it has been demonstrated that the mechanical metal-cutting industry can achieve significant productivity gains by implementing tool condition monitoring. This paper reviews a study on fault diagnosis of drilling tools by applying a data-driven approach. The diagnosis of drilling-tool faults involves the measurement of various parameters using different sensors. The data acquisition system must adapt to the sensor and analyze the data depicted in writing. Various data extraction techniques are available, including statistical analysis, histograms, and wavelet analysis. A wide variety of deep learning and machine learning techniques can be applied in data-driven approaches to identify drill tool failures. Owing to the different algorithms, diagnostic tests were carried out to identify drilling tool failures, which are also presented in the writing. The experiment involved using the drilling tool at various stages of wear, that is, in both new and faulty conditions using a CNC drilling machine, and obtaining vibration data in each of these scenarios. The vibration signals were captured using a piezoelectric accelerometer. The statistical features were extracted using these vibration signals. Drilling tool fault classification was performed using the best-first tree classifier. The classification accuracy obtained by this algorithm was 96.2264%. A data driven intelligent approach can be used in future study to diagnose the fault characterization of the drilling tool.

Modeling Lymphatic Filariasis by incorporating disease awareness through fractional derivative operators.