Current Issue [Vol. 11, No. 12] [December 2025]

The present study evaluates the effectiveness of natural seed priming agents—banana peel water, onion extract, aloe vera water, cow urine, and normal water—on the germination and early seedling vigor of five Indian wheat (Triticum aestivum) varieties: Malavraj HI 8737, Shree Ram Super 303, Pusa Purna 1544, HD 3385, and Lokwan. The objective was to explore eco-friendly, low-cost alternatives to synthetic seed treatments to enhance seed performance and support sustainable agricultural practices. Seeds from each variety were soaked for 8 hours in the respective priming solutions, air-dried, and placed in Petri dishes lined with moist tissue paper. Germination parameters—including germination percentage, speed, root and shoot length, and seedling vigor index—were recorded daily over a 10-day period under controlled conditions. Significant differences were observed across treatments and varieties.
Results revealed that Pusa Purna 1544 and Lokwan responded most favorably, particularly under banana peel water, onion extract, and aloe vera treatments, which led to early germination and higher seedling vigor. Malavraj HI 8737 showed the least responsiveness, especially under cow urine and control treatments. Among the treatments, banana peel water was the most effective, followed by aloe vera water, onion extract, normal water, and cow urine.
In conclusion, natural priming agents such as banana peel water, aloe vera water, and onion extract significantly enhance germination and early growth in certain wheat varieties. These treatments present viable, sustainable alternatives for smallholder and organic farmers. Further field-level studies are necessary to validate these findings across diverse agro-climatic zones.
Keywords: Seed priming, Natural priming agents, Germination, Seedlingvigor, Banana peel water, Onion extract, Aloe vera water, Cowurine, Sustainable agriculture, Organic seed treatment.

Federated learning (FL) enables collaborative model training across distributed clients while preserving data privacy by avoiding raw data transmission. However, FL systems face critical challenges in balancing three competing objectives: privacy protection, communication efficiency, and model accuracy. This comprehensive survey reviews the state-of-the-art methods and mechanisms designed to address these challenges through privacy-aware compression. We systematically reviewed 50 seminal papers published between 2018–2025 across leading venues using a structured search methodology. The surveyed approaches are organized into four primary families: (i) numerical mechanism design approaches (e.g., Minimum Variance Unbiased and Interpolated MVU mechanisms) that jointly optimize noise allocation and bit budgets; (ii) compression-based methods (quantisation, sparsification, compressive sensing, and learned auto encoders) with privacy guarantees; (iii) incentive-compatible game-theoretic frameworks (e.g., auction and contract theory) for personalised privacy budgets; and (iv) real-world implementations on edge devices and IoT systems. Empirical evidence demonstrates that well-designed combinations of these approaches can achieve communication reductions of 50–90% while maintaining model accuracy within 1–3% of non-private baselines under formal privacy constraints (e.g., ε ≤ 2 differential privacy). We identify persistent challenges including non-linear privacy-utility-communication trade-offs, robustness to adaptive adversaries, handling client heterogeneity, and deployment feasibility. Finally, we discuss open research directions including Byzantine-robust compression, real-world regulatory compliance, and adaptive online compression schemes. This survey provides a unified perspective on how mechanism design, cryptographic, and learning-theoretic techniques can be synergistically combined to enable practical, provably private, and communication-efficient federated learning systems.
Keywords: adaptive quantization, compressive sensing, differential privacy, federated learning, incentive mechanisms, privacy-aware compression

The article is dedicated to the analysis of modern approaches and design patterns for high-load fault-tolerant systems. The main problems such as scalability, availability, and data consistency in distributed systems are considered, along with solutions through microservice architectures, event-driven systems, and containerization. The advantages and disadvantages of various architectural solutions, including patterns like Circuit Breaker and data replication, are assessed. The evolution of web application architecture is also discussed, from monolithic solutions to modern serverless architectures. The research helps to understand the key factors influencing the design of fault-tolerant systems and identifies promising development directions in this field.
Keywords: High-load systems, fault tolerance, microservices, scalability, containerization

Grinding is a widely employed machining technique for producing components requiring high dimensional and surface accuracy in mechanical manufacturing. This study performs a multi-objective optimization of the surface grinding process for SCM400 steel using a surface grinding machine. A total of nine experiments were designed based on the Taguchi approach. In each experiment, three cutting parameters—workpiece velocity, feed rate, and depth of cut—were varied. Four response variables, namely surface roughness (Ra), and the cutting force components in the x- (Fx), y- (Fy), and z-directions (Fz), were measured.
The ENTROPY method was adopted to determine the weighting coefficients of the performance criteria, while the Root Assessment Method (RAM) was employed to solve the multi-objective optimization problem. The results indicate that the optimal workpiece velocity, feed rate, and depth of cut are 10 m/min, 4 mm/stroke, and 0.01 mm, respectively. Under these optimal conditions, the corresponding values of Ra, Fx, Fy, and Fz are 0.49 μm, 18.4 N, 15.2 N, and 28.4 N.
Keywords: surface grinding, SCM400 steel, multi-objective optimization, Entropy method, RAM method.

This community service initiative aims to address the challenges in mathematics education by introducing an innovative approach through the establishment of the School Mathematics Health Center, utilizing a learning environment for tutoring in the city square and "go to school” for teachers in remote villages. This research employs a descriptive design, involving 40 teachers, 20 early childhood educators, and 100 elementary school students. Data were collected through structured interviews, observations, and documentation, and analyzed using triangulation and data reduction techniques. The findings suggest that the new educational materials used at the Health Center significantly enhanced student engagement and comprehension of mathematical concepts. Participants showed marked improvements in their attitudes toward mathematics, becoming more enthusiastic, cheerful, and confident. The individualized approach also contributed to better focus and concentration during lessons. Parents’ positive feedback underscores the program’s effectiveness in improving academic outcomes and providing an affordable educational solution. This initiative highlights the potential of open-space learning environments to foster a deeper love for mathematics and enrich the overall educational experience for both ECE and elementary students. Teachers in remote village schools have increasingly developed knowledge of the latest issues in mathematics education and have innovative ideas for implementing ethnomathematics teaching directly in the classroom.
Keywords: free tutoring, city square, mathematics learning, remote villages

This study examined how heat-treatment conditions influence the equilibrium moisture content, water-repellent efficiency, anti-swelling efficiency, Brinell hardness, modulus of rupture, and modulus of elasticity of Cinnamomum cassia wood treated in a mixed oil medium composed of natural oleoresin and waste cooking oil. The oleoresin, a non-timber forest product obtained from Dipterocarp species, was combined with recycled cooking oil collected from an oil recovery facility. Heat treatment was conducted using an oleoresin–waste cooking oil mixture at a mass ratio of 20:80, with treatment temperatures between 130 and 170 °C and durations ranging from 90 to 180 min. Response surface methodology was applied to model the relationships between process variables and wood properties, resulting in four linear and two quadratic models with high adjusted coefficients of determination, indicating good model reliability. The findings revealed marked enhancements in the dimensional stability of Cinnamomum cassia wood, while treatment parameters significantly affected its mechanical performance, allowing optimal heat-treatment conditions using the oleoresin–waste cooking oil system to be determined.
Keywords: Cinnamomum cassia wood, Heat treatment, Oleoresin, Response surface methodology, Waste cooking oil

Optimization of Machining Process Parameters for 90CrSi Steel Grinding.
Optimization of the machining process plays a crucial role and significantly affects the efficiency of the entire machining operation. This study applies a probability-based method to optimize the grinding process for 90CrSi steel material.
The experimental process was carried out with nine experiments designed according to the Taguchi matrix (or Taguchi orthogonal array). In each experiment, the values of the cutting parameters (or machining parameters) were varied: workpiece velocity (or work speed), feed rate (or depth of stroke), and depth of cut.
Four responses (or output parameters) were measured in each experiment: surface roughness (Ra), cutting force component in the X-direction ($F_x$), cutting force component in the Y-direction ($F_y$), and cutting force component in the Z-direction ($F_z$).
Two methods, MEREC (Method based on the removal effects of criteria) and SPC (Sequential Programming Criteria) (or another suitable Multi-Criteria Decision Making - MCDM method, depending on the full context of the acronym 'SPC'), were utilized to calculate the weights for the responses. Subsequently, the Probability method was employed to rank the experiments.
The results showed that the experiment corresponding to the cutting parameter values of workpiece velocity $= 10$ (m/min), feed rate $= 3$ (mm/stroke), and depth of cut $= 0.01$ (mm) was the best among the conducted experiments. These values represent the optimal cutting parameters.
Corresponding to these optimal cutting parameter values, the surface roughness ($Ra$), and the cutting force components ($F_x, F_y, F_z$) were $0.2772$ ($\mu m$), $85.0080$ (N), $55.3280$ (N), and $112.6912$ (N), respectively.
Keywords: Grinding, 90CrSi steel, MCDM (Multi-Criteria Decision Making), Probability, weight method

Synclinal bottoms represent structurally complex zones that contain significant coal resources but also pose major technical and safety challenges for underground coal mining. In the Quang Ninh coal basin (Vietnam), many underground mines are currently approaching synclinal bottoms under conditions of increasing mining depth, complex geology, and limited mechanization. However, systematic assessments of the impacts of syncline geometry on mining organization and technical solutions remain limited. This study investigates the characteristics and mining implications of synclinal bottoms in typical underground coal mines in the Quang Ninh region, with a focus on seam geometry, hydrogeological conditions, gas accumulation, and ground stability. A combined methodology integrating geological analysis, field investigation, and comparative assessment was applied to classify synclinal bottoms into bowl-shaped and trough-shaped types. Their influences on mine opening systems, panel layout, ventilation, drainage, and mining efficiency were evaluated. The results indicate that synclinal bottoms significantly differ from seam limbs in terms of seam dip variability, water and gas concentration, and strata behavior, leading to increased difficulties in longwall arrangement, drainage, and ventilation. Mining performance is strongly controlled by syncline geometry and opening schemes. Synclines developed from the bottom upward with gravity drainage show better safety and recovery efficiency, whereas closed or deeply seated synclinal bottoms are associated with higher coal losses and operational risks. The study emphasizes that synclinal bottoms should be treated as specialized mining units rather than extensions of seam limbs. The findings provide a scientific basis for selecting appropriate mining solutions for different syncline types and contribute to safer and more efficient underground coal mining in the Quang Ninh coal basin and other structurally complex coalfields.
Keywords: Synclinal bottom, mine ventilation, mine drainage, underground coal mining, water accumulation

The study is devoted to the systematization and analytical examination of architectural patterns of multi-agent systems (MAS) aimed at automating end-to-end corporate business processes. The relevance is driven by a shift from monolithic AI models to flexible ensembles of specialized autonomous agents that jointly solve complex tasks, while the lack of established design practices hinders the construction of reliable and scalable solutions. The scientific novelty lies in proposing a taxonomy of MAS patterns — hierarchical, collaborative, and sequential — and in developing an original framework that enables the selection of an optimal pattern given the characteristics of the target business process. The work considers both classical and modern approaches to agent design and elucidates key aspects of their interaction: communication, coordination, and state management. Particular attention within the study is devoted to hybrid architectures as the most realistic option for complex corporate landscapes. The goal is to form a methodological foundation for MAS design in the corporate environment; to this end, pattern analysis, systems analysis, and conceptual modeling are employed. In conclusion, a decision-making framework is presented that helps architects align process requirements with the type of architecture. The material is intended for IT architects, engineers, and project managers in the field of AI automation.
Keywords: multi-agent systems, architectural patterns, corporate business processes, AI agents, LLM agents, automation, systems architecture, AutoGen, CrewAI, design pattern

Solving multi-objective optimization problems is a fundamental task across various fields, including economics, management, and engineering. Two critical components of addressing these problems are determining objective weights and selecting an appropriate mathematical method for optimization. This study integrates three methods—Entropy, MEREC, and TOPSIS—to solve a multi-objective optimization problem aimed at identifying the optimal copper-core electrical wire among 28 available alternatives. Specifically, the Entropy and MEREC methods were employed to calculate the criteria weights, while the TOPSIS method was utilized to rank the copper-core wire types.
The results indicate that although the rankings of the electrical wires were inconsistent when using weights derived from different methods (Entropy vs. MEREC), certain alternatives maintained a consistent rank regardless of the weighting approach. The wire type associated with product code 20255114 was identified as the optimal choice among the 28 types evaluated in this study.
Keywords: optimization, weight method, Entropy method, MEREC method, TOPSIS method.

This study investigates the key product attributes that significantly influence consumer buying behaviour towards organic food. Specifically, it examines the impact of five key factors: taste, availability, premium price, social media influence, and packaging and labeling. Five hypotheses were formulated to test the relationship between these attributes and consumer buying behaviour. A close ended questionnaire was created and subsequently distributed among consumers among 320, out of which 284 responses were positively collected. The data were tested with the help of exploratory factor analysis using SPSS 25.Results indicate that taste, availability, social media impact, and packaging & labeling have an important influence on the purchasing decision of customers, whereas premium price is not a determining factor. The research is helpful for companies to formulate efficient marketing strategies and enhance product offerings to cater to the increasing demand for organic products.
Keywords: Buying behaviour, Organic Food, Product Attributes, consumers’ preferences.