This study proposes a hybrid predictive maintenance framework that integrates the Kolmogorov-Arnold Network (KAN) with Short-Time Fourier Transform (STFT) for intelligent fault diagnosis in industrial rotating machinery. The method is designed to address challenges posed by non-linear and non-stationary vibration signals under varying operational conditions. Experimental validation using the FALEX multispecimen test bench demonstrated a high classification accuracy of 97.5%, outperforming traditional models such as SVM, Random Forest, and XGBoost. The approach maintained robust performance across dynamic load scenarios and noisy environments, with precision and recall exceeding 95%. Key contributions include a hardware-accelerated K

Abstract

The use of modern scientific methods and techniques, is considered important topics to solve many of the problems which face some sector, including industrial, service and health. The researcher always intends to use modern methods characterized by accuracy, clarity and speed to reach the optimal solution and be easy at the same time in terms of understanding and application.

the research presented this comparison between the two methods of solution for linear fractional programming models which are linear transformation for Charnas & Cooper , and denominator function restriction method through applied on the oil heaters and gas cookers plant , where the show after reac

In this paper, the problem of resource allocation at Al-Raji Company for soft drinks and juices was studied. The company produces several types of tasks to produce juices and soft drinks, which need machines to accomplish these tasks, as it has 6 machines that want to allocate to 4 different tasks to accomplish these tasks. The machines assigned to each task are subject to failure, as these machines are repaired to participate again in the production process. From past records of the company, the probability of failure machines at each task was calculated depending on company data information. Also, the time required for each machine to complete each task was recorded. The aim of this paper is to determine the minimum expected ti

A new adsorbent was developed by integrating algae biomass (AG) into a chitosan (CN) matrix, followed by structural enhancement via crosslinking with pyromellitic dianhydride (PMDA) through a hydrothermal synthesis approach. This process resulted in the formation of a robust AG@CN-PMDA composite with improved physicochemical characteristics suitable for advanced adsorption applications. The AG@CN-PMDA composite was evaluated for its efficiency in removal of the cationic dye methyl violet 2B (MV 2B) from aqueous solution. The adsorption process was refined through the Box-Behnken design (RSM-BBD), evaluating three essential parameters: adsorbent dosage (A: 0.02–0.1 g/100 mL), pH (B: 4–10), and time (C: 5–20 min). The ideal conditions f

Survival analysis is widely applied in data describing for the life time of item until the occurrence of an event of interest such as death or another event of understudy . The purpose of this paper is to use the dynamic approach in the deep learning neural network method, where in this method a dynamic neural network that suits the nature of discrete survival data and time varying effect. This neural network is based on the Levenberg-Marquardt (L-M) algorithm in training, and the method is called Proposed Dynamic Artificial Neural Network (PDANN). Then a comparison was made with another method that depends entirely on the Bayes methodology is called Maximum A Posterior (MAP) method. This method was carried out using numerical algorithms re

Beyond the immediate content of speech, the voice can provide rich information about a speaker's demographics, including age and gender. Estimating a speaker's age and gender offers a wide range of applications, spanning from voice forensic analysis to personalized advertising, healthcare monitoring, and human-computer interaction. However, pinpointing precise age remains intricate due to age ambiguity. Specifically, utterances from individuals at adjacent ages are frequently indistinguishable. Addressing this, we propose a novel, end-to-end approach that deploys Mozilla's Common Voice dataset to transform raw audio into high-quality feature representations using Wav2Vec2.0 embeddings. These are then channeled into our self-attentio