Support vector machines (SVMs) are supervised learning models that analyze data for classification or regression. For classification, SVM is widely used by selecting an optimal hyperplane that separates two classes. SVM has very good accuracy and extremally robust comparing with some other classification methods such as logistics linear regression, random forest, k-nearest neighbor and naïve model. However, working with large datasets can cause many problems such as time-consuming and inefficient results. In this paper, the SVM has been modified by using a stochastic Gradient descent process. The modified method, stochastic gradient descent SVM (SGD-SVM), checked by using two simulation datasets. Since the classification of different ca

This study focused on treatment of real wastewater rejected from leather industry in Al-Nahrawan city in Iraq by Electrocoagulation (EC) process followed by Reverse Osmosis (RO) process. The successive treatment was applied due to high concentration of Cr3+ ions (about 1600 ppm) rejected in wastewater of this industry and for applying EC with moderate power consumption and better results of produced water. In Electrocoagulation process (EC), the effect of NaCl concentration (1.5, 3 g/l), current density (C.D.) (15-25 mA/cm2), electrolysis time (1-2 h), and distance between electrodes (E.D.) (1-2 cm) were examined in a batch cell by implementing Taguchi experimental design. According to the results obtained from multiple regression and signa

Drones play a vital role in the fundamental aspects of Industry 4.0 by converting conventional warehouses into intelligent ones, particularly in the realm of barcode scanning. Various potential issues frequently arise during barcode scanning by drones, specifically when the drone camera has difficulty obtaining distinct images due to certain factors, such as distance, capturing the image whilst flying, noise in the environment and different barcode dimensions. In addressing these challenges, this study proposes an approach that combines a proportional–integral–derivative (PID) controller with image processing techniques. The PID controller is responsible for continuously monitoring the camera’s input, detecting the difference

Manufacturing industry is constantly looking for ways to improve the efficiency of production processes so as to lower production costs, maintain product quality, and improve production planning. This study aims to provide an integrated optimization and forecasting framework that combines Cutting Optimization Pro5, linear programming, and an Enhanced N-BEATS deep learning model to minimize material waste and manufacturing cost and to predict the number of production items with a price-reduction strategy. The real industrial case study is based on production and pricing data from the Akad factory of the General Company for Electrical and Electronic Products in Iraq for the period from 2014 to 2025, concerning the manufacture of 80 L

The present study explores the solar-induced photocatalytic degradation of reactive red (RR) and reactive turquoise (RT) dyes in a single system using TiO2 immobilized in xanthan gum (TiO2/XG), synthesized using the sol–gel dip-coating technique for direct precipitation. SEM-EDX, XRD, FTIR, and UV–Vis were used to assess the characteristics of the resulting catalyst. Moreover, the effects of different operating parameters, specifically pH, dye concentration, TiO2/XG concentration, H2O2 concentration, and contact time, were also investigated in a batch photocatalytic reactor. The immobilized TiO2/XG catalyst showed a slight adsorption degradation efficiency and then improved the RR and RT dye degradation activity (92.5 and 90.8%