INTRODUCTION: Drinking water that is clean and safe is important for everyone's health. About 1.4 million deaths worldwide are noted to contaminated drinking water each year. Because contaminated water sources are the primary cause of diarrheal infections, they account for about 505,000 deaths every year. To overcome these challenges, this work proposes an integrated IoT and AI-based solution for real-time, multi-nutrient water quality analysis.

OBJECTIVE: In this paper, our objective is to develop a complete system which is integrated with IoT-based water nutrient analysis system using advanced machine learning models that can predict multiple nutrient levels for better crop. To increase the interpretability, reliability, and security of water quality monitoring system.

MATERIAL/METHOD: For data collection we deployed the IoT sensors in different sources like reservoirs, irrigation canals, and ponds for continuously monitoring the parameters like:- phosphorus (P), potassium (K), pH, Temperature, BOD etc. The data which we have collected from the sensors are securely transmitted to a cloud-based platform using end-to-end encryption protocols. Advance machine learning classifiers ensemble learning algorithms are used to analyze the real-time data to gives multi-nutrient predictions. The dataset was collected from GIETU agricultural fields over 6 months from2024 January to till date. We also used Explainable AI (XAI) techniques interpret properly of the machine learning algorithms.

Result: The performance metrics like accuracy, precision, recall, and F1-score are calculated for predicting the water quality. Our experimental observation reveals that the ensemble classifier RFS (Random Forest + SVM) classifier exhibits well and having an accuracy of 90% as comparison to other models. The hybrid classifier significantly higher than the traditional approaches .As well as we used XAI techniques to increased the interpretability of the classifiers to make effective decision-making for water management. For data security we used encryption and decryption algorithms ensured data integrity and protection against unauthorized access.

The beer industry is a significant global market, and the number of small breweries is on the rise. Centrifugal pumps are vital to the production system, especially in microbreweries. Failures in the machine setup, such as incorrect valve positions or blockages in the piping system, can result in increased production time and energy consumption, the need for rework, and even impact the quality of the beer. In this context, the correct operation of the pump can be monitored, and faults detected early through Intelligent Fault Diagnosis (IFD). Since microbreweries typically have a low level of automation and limited resources for investment, this article aims to develop an IFD capable of detecting blockages and inlet problems in centrifugal pumps by exploring a soft sensor approach. The predictive model is constructed based on data provided by centrifugal pump drives, such as current, torque, and power factor, and does not require additional sensors. This data is collected through a managed switch with a mirrored port, captured with Wireshark, and interpreted by a Python script that extracts univariate statistical features. A dataset containing 1260 samples was created, covering data from healthy operation, closed pump inlet, and closed pump outlet. With this dataset, predictive models were trained using SVM and ANN to identify the pump's operating condition. Performance analysis was conducted using accuracy and precision metrics. The results of this study are promising, indicating that the proposed IFD approach can be effectively utilized in developing sensors specifically designed for Industry 4.0 applications. This approach enhances automation and predictive maintenance capabilities, making it particularly beneficial for small-scale breweries looking to optimize their production processes and maintain high-quality standards.

With the rapid rise in technological advancements, people's health can be tracked and monitored in multiple ways. This enables them to focus more on healthier lifestyles minimizing health issues concerning long screen time. Artificial Intelligence (AI) techniques like Large Language Model (LLM) technology enable intelligent smart assistants to be used on mobile devices and other cases. The proposed system uses the power of LLMs and creates a virtual personal assistant for long-time screen users by monitoring their health parameters with various sensors for real-time monitoring of the seating posture, heartbeat, stress level, motion tracking and eye movement etc., to constantly track and give necessary advice and making sure that their vitals are expected and in safety parameters. The intelligent system combines the power of AI and Natural Language Processing (NLP) to build a virtual assistant embedded into the screens of mobile devices, laptops, desktops and other screen devices, which employees across the various workspaces use. The intelligent screen, with the integration to various sensors, tracks and monitors the users' vitals along with various other required health parameters and alerts them to take breaks, have water, refresh and ensures that the users stay healthy while using the system for work. These systems also suggest the required exercises for the eyes, head and other body parts. The proposed system being supported with the possible users' databases, recognize the current user and suggest the advisory accordingly. The system also adapts and ensures the users get proper relaxation and focus when using the system. The intelligent screen system monitors and improves the health of the employees who have to work for a long time, thereby enhancing the productivity and concentration of the employees in various organizations.

Agriculture is a country's backbone and plays an essential role in shaping its economic performance. Factors like disasters, extreme weather changes, pests, and soil quality greatly impact’s productivity resulting in economic loss. Accurate predictions in agricultural practices, such as crop recommendations, can significantly enhance productivity and resource management. The objective of this research is to build a robust crop recommendation system using an ensemble model composed of several regression machine learning (ML) models. The study uses a real-time dataset collected using IoT sensors for crop recommendation of 22 different crop varieties. The dataset is available in Kaggle. The main features of the dataset are nitrogen, phosphorus, potassium and Ph value of soil, humidity and rainfall. This study compares the performance of regression models with ensemble models and the impact of different hyperparameter tuning techniques such as Bayesian Optimization, Genetic algorithm and GridSearchCV. Fune-tuning is done to improve the predictive performance thus providing smarter agriculture. Regression Models like Logistic Regression, SVM, Decision Tree, Navie Bayes, K Nearest Neighbour, Extra Tree Classifier, XGBoost, Gradient Boost models are compared with ensemble models like voting, bagging, boosting and stacking ensemble model. Among all the ensemble techniques stacking ensemble obtained a highest accuracy of 99.3% when compared to other regression models.

Flooding is an issue that affects many cities during periods of heavy rain, especially in developing countries. This issue often happens due to the lack of timely maintenance and cleaning of rainwater drains. This work presents a proposal to detect blockages in rainwater drains to determine where maintenance and cleaning need to be done promptly. An architecture that includes sensor nodes, a Gateway, and a cloud application was defined. The sensor nodes, which detect potential blockages in the drains, send data and status signals to the Gateway using LoRaWAN. The Gateway then passes the data to a cloud platform that records the data and issues an alert when a blockage is detected in a drain. The sensor node prototype is based on a Heltec LoRa Wifi board and has two sensors: an HC-SR04 ultrasonic sensor to measure distance and a DHT22 sensor to measure humidity and temperature. The cloud application was developed on the Arduino Cloud platform. A decision tree is proposed to detect blockages based on the readings from these sensors, particularly the distance sensor, considering four possible states: clear, possibly blocked, potentially blocked, and blocked. The decision tree is implemented in each node. Periodically, the node collects data from the sensors and transmits it to the cloud, and if a potential blockage is detected, it sends a message that triggers an alert on the platform. Preliminary tests of the prototype show accurate and timely results in detecting potential blockages, allowing for better use of resources allocated for maintaining drainage systems.

Introduction: A Wireless Sensor Network (WSN) consists of sensors with wireless transceivers that link autonomously over many hops. It offers various advantages, including less traffic, more stability, extended wireless communication distances, and broader coverage regions at less money. Combining emerging Limited Bandwidth Channel Internet of Things (LBC-IoT) technologies with wireless sensor networks offers interesting applications in defence, medical, smart transportation, and commercial sectors.

Methods: This study initially analyses WSN and LBC-IoT technologies independently before combining them to look into the networking framework of LBC-IoT and WSN, as well as the associated technologies resulting from the fusion. The article describes the typical network node redeployment strategy for wireless sensors, which can lead to poor node connection and inadequate coverage due to a lack of local subgroup node search. The proposed WSN node localization technique, based on the Hop Count Path Vector (HOP-PV) algorithm, enhances and optimizes the calculation method of the number of node hops and average hop distance, resulting in the PVHOP-LBCIOT mechanism.

Results and Discussion: Simulation results indicate that the improved PVHOP-LBCIOT algorithm's three deployment methods (square, central uniform, and cross) outperform the two methods of HOP-PV (random deployment) and PVHOP-LBCIOT (border uniform deployment) for an equal number of unknown moving anchor positions (11), unequal number of unknown nodes (30-13)and fixed communication radius (6), with reduced average error rate of 32.79 % from 38% and improved accuracy for obtaining unknown node location.

Conclusion: The suggested WSN node localization approach using a single mobile anchor node, known as the PVHOP-LBCIOT mechanism, enhances and optimizes the process of calculating the average hop distance and the number of node hops. A comparison experiment demonstrates that this hopping algorithm has much greater coverage, node power, connectivity, and robustness compared to traditional method.

In recent years, researchers developed the concept of Construction 4.0, which seeks to integrate Industry 4.0 technologies within the construction sector; one of the critical technologies is the use of robots; thus, different studies have documented the design of robots for certain specific tasks achieving benefits such as improved productivity and safety. However, the literature is scarce on strategies for implementing robots in companies under a specific and local context, especially in developing countries where barriers limit their adoption. The following study proposes a roadmap to initiate the implementation of robots in construction companies in developing countries, taking the Peruvian construction sector as a case study. Based on previous research developed by the authors, the first stage identified barriers and benefits of adopting robots in the Peruvian context. After that, the authors proposed a roadmap based on previous studies, and with a proposal, Peruvian 6 experts validated the roadmap in two rounds to see its feasibility. As a result, the proposal roadmap consists of four phases divided into 11 processes: phase 01, which includes the alignment of the company with the need to implement robots; Phase 02, where the company evaluates the technology and construction process to implement robots; Phase 03, where the company plans the implementation in construction site; and phase 04, where the company implements robots and evaluates good practices and opportunities for improvement. The study's novelty consists of proposing steps that involve adopting robots from conception to project closure in developing countries, where adopting this type of technology is still limited due to factors such as lack of knowledge of the benefits, among others. The following study contributes to professionals and companies in the construction sector who wish to implement robots in their projects, maximize the benefits of implementation, and minimize the risks of adoption.

Effective weed detection is essential in modern agriculture to improve crop yield and quality. Farmers can optimize their weed control strategies by applying tailored herbicides based on accurate identification of weed species and the areas they affect. Real-time object detection has been transformed by recent advances in image detection technology, especially the YOLO (You Only Look Once) algorithm, of which YOLOv7 has shown to be more accurate than its predecessors in weed detection. Because of its novel E-ELAN layer, the YOLOv7 model achieves an astonishing 97% accuracy, compared to the estimated 78% accuracy of the YOLOv5 model. cropThis study suggests using Internet of Things (IoT) sensors in conjunction with YOLOv7 to improve weed detection using an integrated strategy. It is advantageous to include a variety of sensors in the proposed work in detecting and managing weeds with greater accuracy and comprehensiveness can be achieved by combining a variety of sensors to improve the data obtained. An enhanced weed detection system can be achieved by utilizing the distinct information that each type of sensor provides. A comprehensive set of environmental data, including soil moisture, temperature and humidity, light intensity, pH, and ultrasonic distance sensors, will be used to correlate with patterns of weed growth. This information will be sent to a central Internet of Things gateway for in-the-moment analysis and merging with video footage taken agricultural fields.
Farmers can anticipate weed infestations and optimize their management tactics thanks to predictive analytics made possible by the integration of sensor data with YOLOv7's weed detecting capabilities. The potential for large herbicide application cost savings and improved crop yields, which would increase farmer profits, highlight the economic viability of this strategy. This methodology seeks to revolutionize weed control procedures by utilizing cutting-edge technology and IoT connectivity, making them more effective and efficient .

Surface charging has been widely used in various functionalization technologies to create (bio)chemical sensing layers. This is due to the unique ability of electrostatic interactions not only to immobilize the desired receptors, but also to control their uniform distribution over the surface. This physical process is widely used in interfacial electrostatic driven reactions, classical layer-by-layer deposition, electrostatic levitation of proteins over the surface. Thiocyanates are extremely promising compounds for creating an ultrathin charged layer on the SPR, QCM, etc. transducers, due to their small size and ability to self-organize into a monolayer on the gold surface. For sensor science, an important issue is the experimental confirmation of the presence of the effective negative charge within or near their monolayer. This study is devoted to clarify this issue.

To determine the sign of the surface charge of gold modified by thiocyanate monolayer, we used 60 nm organic stabilized silver nanoparticles as an electrostatic probe: in distilled water, the charge of these compounds changes from the strongly positively charged Ag-NP&BPEI, through Ag-NP&PEG and Ag-NP&PVP with a moderate negative charge, to the highly negatively charged Ag-NP&CIT.

Analysis of SPR and UV-VIS spectroscopy results, electrochemical measurements and wide-field surface plasmon resonance microscopy imaging indicate that gold modified with thiocyanate has the maximum adsorption capacity for objects with a positive charge on their shell. Strongly negatively charged nanoparticles are not adsorbed on a surface modified with thiocyanate. Visualization of nanoparticles using wide-field Surface Plasmon Resonance Microscopy confirmed the sticky adsorption from suspension of Ag-NP&BPEI nanoparticles on the surface of gold modified with a self-assembled thiocyanate layer.

The authors express their gratitude to Professor Vladimir Mirsky for support in the research and fruitful discussions. This work has received funding through the MSCA4Ukraine project (Grant ID 1119494), which is funded by the European Union.

Tool Condition Monitoring (TCM) systems have become increasingly important in industrial automation due to the need to improve efficiency and reduce manufacturing costs. These systems use advanced sensors to capture signals during machining processes, allowing for early detection of faults and prediction of tool life. This study explores the potential of using the Cosine Similarity (CS) method as a practical technique for analyzing acoustic emission (AE) signals and monitoring tool wear during milling operations. Acoustic signals were applied to the CS method under reference conditions and after potential damage. We used 9000 samples of the milling cutter passing over the workpiece, collected from experiments with milling machines using the AE sensor WD925 at a frequency of 100 kHz. The CS method tracked wear proportionally in each case. As the tool wore down, its similarity to the intact tool decreased, proving to be an effective indicator for condition monitoring. However, the change in CS calculation was not as pronounced as the tool wear observed, suggesting that having a sufficient amount of data is crucial for this methodology in condition monitoring. A longer sampling period is necessary to capture significant signal variations and effectively detect losses in similarity. This provides a significant amount of data and, as a result, leads to more conclusive findings for the process in question.