Volume 8.
Issue 06.

Mechanical impurities contained in mine water negatively affect the operating process of centrifugal pumps. Mine water contains solid particles of various sizes and concentrations, such as fragments of rock, quartz, and granite, which cause hydro-abrasive, corrosive, and cavitation wear of the pump working elements. As a result, the efficiency of the pump decreases, the pressure head drops, and energy consumption increases.

A large number of organizations have been increasingly adopting hybrid cloud computing models due to the significant acceleration of digital business transformations. These hybrid clouds offer organizations scalability and flexibility from public cloud platforms combined with control and compliance from their on-premises infrastructure. Hybrid cloud models also present organizations with several advantages such as greater operational agility; lower costs; and greater workload scalability. However, hybrid clouds create a variety of new and complex cybersecurity issues related to identity management; cross-cloud visibility; data governance; regulatory compliance; and distributed attack surfaces. The objective of this research is to analyze and compare several of the most recent data security methods used by hybrid cloud systems for digital businesses. In order to achieve the objective, the research will employ a multi-faceted approach consisting of literature synthesis; technical framework analysis; a metric based assessment model; and a simulated threat modeling process. Advanced security methods such as Zero Trust Architecture (ZTA); Cloud Security Posture Management (CSPM); Cloud Workload Protection Platforms (CWPP); Homomorphic Encryption; Secure Multi-Party Computation (SMPC); Trusted Execution Environments (TEEs) and AI-driven Security Operations (SecOps) were selected for examination. The proposed evaluation method examines each identified technique using a set of multiple operational and security metrics such as confidentiality; integrity; availability; scalability; compliance alignment; performance overhead and incident response efficiency. Evaluation results indicate that organizations can increase their hybrid cloud ecosystem's security resilience through the use of integrated security frameworks incorporating elements of ZTA; automated CSPM/CWPP components and hardware-enforced trust technologies. The proposed framework can assist decision makers at organizations in determining how best to implement their hybrid cloud security solutions in accordance with current cybersecurity regulations and standards.

The paper presents a scientifically grounded composition-design approach for non-autoclaved cement–fly ash foam concrete modified with wollastonite and basalt fiber. The study is focused on the stabilization of the cellular structure, strengthening of interpore partitions and improvement of the balance between thermal insulation and mechanical performance. A two-stage preparation route was adopted: technical foam was produced separately and then introduced into a cement–fly ash matrix containing dispersed wollastonite and, when required, basalt fiber. The main controlled parameters were water-to-solid ratio, cement-to-fly ash ratio, foam density, wollastonite fraction and dosage, wollastonite fineness, and fiber content. The results indicate that dispersed wollastonite at 1.0–1.5% of binder mass reduces fresh-mixture settlement from 14 mm in the control mixture to about 6 mm and increases the foam stability index to 0.94–0.95 at a specific surface of approximately 300 m²/kg. The most effective method of addition was dry blending with cement and fly ash, which produced a dry density of 532 kg/m³, thermal conductivity of 0.128 W/(m·°C), and 28-day compressive strength of 1.92 MPa. Compared with the control mixture, the combined dispersed-wollastonite and basalt-fiber modification increased 28-day compressive strength by about 48%, reduced drying shrinkage by about 27%, and increased the closed-pore fraction from 46% to 52%.

Simulation technologies are becoming an increasingly important component of railway transport safety systems due to the growing complexity of operational processes and the continuing influence of the human factor on traffic safety. This study examines international approaches to the use of simulator technologies in locomotive crew training and analyses their role in improving railway safety. The research is based on a comparative analysis of centralized and decentralized training models used in the Russian Federation, the United States, and European countries. The study reviews modern simulator systems, including full-scale simulators, procedural simulators, scenario-based simulation technologies, and digital learning environments. Particular attention is paid to the integration of simulator technologies into safety management systems, personnel assessment, emergency preparedness, and human factor research. Based on the analysis conducted, a conceptual model for the integration of simulator technologies into railway safety systems is proposed. The results demonstrate that advanced simulation technologies not only improve professional training quality but also contribute to accident prevention, operational risk reduction, and the development of sustainable decision-making skills under complex operating conditions. The study concludes that the further development of simulation technologies is associated with the expansion of their analytical capabilities and deeper integration into comprehensive railway safety management systems.

This paper assesses ambient-cured one-part geopolymer concrete in which powdered sodium metasilicate is used as the principal alkaline activator. The material concept is based on a dry binder system where class F fly ash, ground granulated blast furnace slag (GGBFS) and solid activator are blended before water is added. Such a procedure reduces the dependence on strongly alkaline liquid solutions and brings the mixing sequence closer to ordinary concrete production. Five one-part geopolymer mixtures were examined and compared with an ordinary Portland cement reference in terms of setting behaviour, workability, compressive strength, flexural strength, splitting tensile strength, modulus of elasticity, water absorption, permeable void volume, drying shrinkage, creep, restrained shrinkage and pore-structure features. The 28-day compressive strength of the one-part mixtures ranged from 26.2 to 41.7 MPa, while flexural and splitting tensile strengths were 4.55-6.35 MPa and 3.65-5.05 MPa, respectively. The mixture containing 60% fly ash and 40% GGBFS, activated only with powdered sodium metasilicate anhydrous, provided the most balanced response. Reducing the water-to-precursor ratio to 0.35 improved strength, microstructural compactness and creep resistance, although it shortened the workable period. Compared with the cement reference, several geopolymer mixtures showed higher water absorption and free drying shrinkage; nevertheless, optimized formulations exhibited lower creep and restrained shrinkage. The findings confirm that powdered sodium metasilicate can be an effective single activator for ambient-cured one-part geopolymer concrete when precursor balance and water dosage are controlled together.

This article examines plaster mixes from ancient times to modern types such as traditional gypsum, cement, and lime mortar types. Since dry construction mixtures did not exist at all until the mid-20th century, this provides an analysis of the development of modern plaster compositions. As the population's standard of living improves, consumers are increasingly paying attention to the issues of environmental safety and the harmlessness of the materials used. Materials for interior finishing have the greatest impact on human health, not only because they are in direct contact with humans but also because they participate in the formation of the room's microclimate.

The high growth rate of digital financial ecosystems has greatly amplified the magnitude, speed, and sophistication of fraudulent transactions that have presented major challenges to the conventional fraud detection systems. The traditional rule-based and statistical models usually have high false positive rates, slow detection, and minimal capability to adjust to changing patterns of fraud. In this study, it is proposed to develop a deep learning-based enterprise risk analytics framework that would allow detecting financial fraud in the real-time and meet the regulatory compliance requirements. The architecture combines sophisticated deep learning models, such as Long Short-Term Memory (LSTM) networks, autoencoders and graph neural networks, with business risk management applications, such as dynamic risk scoring, anomaly detection pipelines, and compliance monitoring layers. The proposed system is tested using the publicly available transactional datasets, like the European card fraud data, on the basis of the most important performance indicators, including accuracy, precision, recall, F1-score, and area under the receiver operating characteristic curve (ROC-AUC). The results show that deep learning models by far exceed traditional machine learning methods by being more accurate in detection and significantly lowering false positives in highly imbalanced datasets. Additionally, explainable AI methods improve model transparency, which can be easily accepted by regulators and audited. The research will add to the body of knowledge by filling in the gap between superior artificial intelligence methods and risk governance models on an enterprise level and provide a flexible and scalable answer to the contemporary financial institutions. The offered framework both enhances the ability to detect fraud and facilitate proactive risk management and compliance in more sophisticated financial settings.

Hair artifacts in dermoscopic images distort the boundaries of skin lesions, alter texture features, and reduce the accuracy of automatic segmentation and classification systems. In this paper, a method based on morphological Black-hat filtering, probabilistic Hough transform, and local median interpolation is proposed to remove hair artifacts from dermoscopic images. The proposed method is computationally simple and CPU-efficient, and it was evaluated using PSNR and SSIM metrics. Experimental results show that this method improves the visual quality of dermoscopic images, removes hair artifacts while preserving important diagnostic features of the dermoscopic image, and is a suitable preprocessing method for the next segmentation step.

This article investigates lightweight porous concrete (LPC) produced from dry mechanically activated mixtures and evaluates the effect of porous aggregates on its physical and mechanical properties. Dry mixtures based on PC500 D0 Portland cement, foam glass, expanded clay, and expanded polystyrene granules were modified using KZ-TM-30 polysilicic acid sol. The nanomodifier improved particle size distribution toward finer fractions, resulting in increased early strength. Compressive and flexural strength, shrinkage, frost resistance, and thermal conductivity of LPC were determined, while microstructural analysis confirmed strong adhesion between the aggregate and cement matrix. The results demonstrate the effectiveness of nanomodified dry mixtures for producing high-performance lightweight concretes with thermal insulation and structural properties.

The rapid expansion of the Internet of Things (IoT) has enabled seamless connectivity among billions of resource-constrained devices, creating new opportunities in smart healthcare, industrial automation, smart cities, and cyber-physical systems. However, this connectivity introduces serious security challenges, particularly in the area of device authentication and secure communication. Mutual authentication is a fundamental requirement to ensure that both communicating entities verify each other before exchanging sensitive data. Traditional authentication mechanisms are often unsuitable for IoT environments due to their computational overhead, energy consumption, and communication latency. This paper proposes a lightweight mutual authentication protocol designed specifically for resource-constrained IoT environments. The proposed framework leverages efficient cryptographic primitives and physical unclonable functions (PUFs) to achieve secure, scalable, and low-overhead authentication. The protocol is evaluated conceptually against common IoT threats such as impersonation attacks, replay attacks, and man-in-the-middle attacks. Comparative analysis with existing state-of-the-art approaches demonstrates that the proposed solution significantly improves efficiency while maintaining strong security guarantees. The study also explores integration scenarios in Internet of Medical Things (IoMT) and edge-enabled IoT architectures.

The article investigates the possibility of using ore and finely dispersed fractions of Koytash wollastonite from the Jizzakh region as a structure-modifying additive in non-autoclaved cement–fly ash foam concrete. The objective of the study was to determine the effect of the fractional state and dosage of wollastonite on the aggregative stability of the mix, the density of interpore partitions, compressive strength, and thermal conductivity. Compared with the control composition, the ore fraction was found to reduce the settlement of the mix from 14 mm to 9–11 mm, while the finely dispersed fraction reduced it to the range of 6–8 mm. When 1.0–1.5% finely dispersed wollastonite was used, the 28-day compressive strength increased from 1.30 MPa to 1.78–1.92 MPa, and the 90-day strength increased to 2.05–2.25 MPa. At the same time, the thermal conductivity coefficient decreased from 0.135 to 0.128–0.130 W/(m·°C). The results were explained by the role of wollastonite as a microfiller, a nucleation center, and a factor promoting the densification of interpore partitions.

Paintless Dent Repair (PDR) technology represents a modern method for restoring vehicle body panels without damaging the original paint coating. In the context of a growing vehicle fleet, increasingly stringent environmental regulations, and rising costs of conventional body repair, the search for resource-efficient and economically viable repair technologies has become particularly relevant. The aim of this study is to provide a comprehensive analysis of PDR technology as an element of the contemporary body repair system from the perspectives of operational efficiency, economic feasibility, and environmental sustainability.

The research employs methods of comparative analysis of conventional and paintless repair technologies, techno-economic assessment of production processes, and synthesis of professional practice data from service centers and insurance organizations. The key technological features of PDR, equipment and personnel qualification requirements, as well as the limitations of the method’s applicability are examined.

Based on the analysis of 124 repair cases (62 PDR and 62 conventional repairs), it was found that the average repair time decreased from 9.4 ± 1.8 hours to 1.7 ± 0.6 hours, while the average cost was reduced from 470 to 185 USD. The annual economic effect at a workload of 35 orders per month amounted to 119,700 USD. The elimination of painting operations made it possible to prevent up to 60.9 kg of volatile organic compound (VOC) emissions per year. The study concludes that integrating PDR technology into modern body repair systems is justified as a sustainable and promising direction that meets economic, environmental, and service quality requirements.

Modular Multilevel Converters (MMCs) have emerged as one of the most significant power electronic converter topologies for medium- and high-voltage applications due to their scalability, modularity, superior output waveform quality, and reduced harmonic distortion. Despite their advantages, conventional MMC control strategies typically depend on continuous capacitor voltage measurement for balancing energy among submodules. Such requirements increase system complexity, sensing costs, communication burden, and susceptibility to measurement inaccuracies. This study presents a comprehensive review and conceptual research framework for an advanced control strategy applied to a 21-level MMC employing Nearest Level Modulation (NLM) without direct capacitor voltage sensing. The proposed framework integrates modulation optimization, circulating current suppression, sensorless balancing mechanisms, and adaptive control principles to achieve stable converter operation while reducing hardware requirements. A detailed synthesis of existing MMC control approaches is conducted, emphasizing modulation methods, predictive control techniques, voltage balancing mechanisms, and renewable-energy-oriented MMC applications. Based on identified research gaps, a sensorless NLM-based control architecture is formulated and analyzed. The framework demonstrates how switching state selection, energy distribution estimation, and current-based balancing algorithms can maintain capacitor voltage equilibrium without dedicated voltage sensors. The study further discusses expected performance improvements in switching losses, harmonic quality, computational efficiency, reliability, and scalability. Findings indicate that sensorless NLM control can provide a practical and economically viable alternative for next-generation MMC systems deployed in smart grids, HVDC transmission, renewable energy integration, and industrial drive applications. The research contributes a structured theoretical foundation for future implementation and validation of advanced MMC control systems.

The paper presents an experimental and analytical study of high-strength lightweight fiber-reinforced concrete designed for hot-dry climatic conditions and for the use of locally available mineral resources. The proposed composite combines CEM I 42.5N Portland cement, silica fume MK-85, limestone powder, ceramic brick waste powder, quartz sand, expanded perlite and dispersed basalt and polypropylene fibers. Expanded perlite was considered not only as a lightweight aggregate but also as a component supporting internal curing through controlled pre-wetting. Four mix variants were compared: a non-fibrous reference mixture, a basalt-fiber mixture, a polypropylene-fiber mixture and a hybrid fiber mixture. The experimental program included tests for fresh concrete workability, hardened density, compressive and flexural strength, water absorption, water tightness, frost resistance, drying shrinkage and thermal conductivity. The hybrid fiber mixture provided the most balanced result: a 28-day compressive strength of 67.9 MPa, flexural strength of 7.97 MPa, hardened density of 1512 kg/m³, shrinkage of 441 µε and thermal conductivity of 0.45 W/(m·K). Regression models confirmed a positive interaction between basalt and polypropylene fibers in strength development and an additional beneficial reduction of shrinkage and thermal conductivity. The scientific novelty consists in the combined use of local expanded perlite, ceramic brick waste and hybrid fiber reinforcement to obtain a lightweight composite with high specific strength, improved crack resistance and enhanced thermal efficiency.

This paper presents the results of a study examining the effect of cocoon stifling and drying methods on fibroin structure. The paper also presents the technological characteristics of studies on raw silk yields dried using various methods. The study concludes that drying cocoons in a microwave field under specific conditions improves cocoon unwinding.

The present study develops and evaluates a non-autoclaved D400 foam concrete modified by a multi-component chemical-mineral system based on a locally applicable GGBFS-type slag, activated natural zeolite, a metakaolin-gypsum complex modifier, an AS setting accelerator and the polycarboxylate ether superplasticizer Sika® ViscoCrete®-1095. The work was aimed at improving the performance of low-density foam concrete without increasing its density beyond the D400 class. Cement paste tests were first used to determine the safe accelerator range by normal consistency and setting time. Fine-grained cement-sand systems were then used to separate the effect of accelerator dosage and water-reducing admixture from the instability of foam. Finally, eight D400 foam concrete compositions were prepared and tested under normal curing and thermal-moisture curing. Strength, density, thermal conductivity, water absorption, softening coefficient, moisture loss, shrinkage deformation, XRD/DTA/TG and SEM indicators were analyzed. The rational composition contained 30% GGBFS-type slag, 8% complex modifier, 0.07% AS accelerator and Sika® ViscoCrete®-1095. Compared with the control D400 composition, the optimized system increased 28-day compressive strength from 2.21 to 4.57 MPa under normal curing and from 1.58 to 3.35 MPa after thermal-moisture curing. Thermal conductivity decreased to 0.073-0.080 W/(m·°C), water absorption decreased from 53.2% to 17.3%, and shrinkage decreased to 0.64-0.68 mm/m. The results confirm that a combined dispersion-packing and early-structure-stabilization approach is effective for producing high-performance non-autoclaved foam concrete from locally adaptable mineral resources.

The increasingly quick growth in the amount, speed, and complexity of cyber threats has placed a great deal of strain on the operational performance of contemporary Security Operations Centers (SOCs), where analysts frequently must deal with a large volume of alerts and have few contextual prioritization tools at their disposal. The rule-based and Security Information and Event Management (SIEM)-based systems though successful in detection often fail to match the threat response with the organizational risk exposure and hence allocate resources optimally and take a long time to respond to incidents. Within the framework of this research, the authors suggest a threat detection approach using machine learning and business impact modeling with the help of AI to offer predictive and context-centered threat prioritization. Based on publicly available cybersecurity datasets, including CICIDS2017 and UNSW-NB15, various supervised learning models, including Random Forest, Extreme Gradient Boosting (XGBoost), and Logistic Regression, were trained and assessed. A quantitative business impact scoring mechanism based on criticality of assets, operational dependency, and potential financial loss was integrated with these models. The hybrid model suggested above produces a composite prioritization score that indicates the probability of a threat and its possible business impact. Empirical evidence shows that the combined method is much more accurate at prioritization, with greater precision and F1-scores than single machine learning models, and lower mean time to detect (MTTD) and mean time to respond (MTTR). The results underscore the importance of inculcating business conditions into AI-based cybersecurity systems. This work is relevant to the development of intelligent SOC design, as it fills the gap between technical threat detection and the organizational risk management by providing a scalable, data-driven solution to improving cyber resilience in complex enterprise settings.

Pharmaceutical and medical device manufacturers operate in a highly regulated environment where operational efficiency, data integrity, and regulatory compliance must coexist. Manufacturing Execution Systems (MES) have traditionally served as the backbone of digital manufacturing, enforcing standardized workflows, batch traceability, electronic records, and compliance with regulations such as FDA 21 CFR Part 11, EU Annex 11, and Good Manufacturing Practice (GMP). While MES solutions provide robust governance and audit readiness, they are often associated with long implementation cycles, high costs, and limited flexibility in addressing shop-floor requirements that change quickly.

Low-Code/No-Code (LCNC) platforms have emerged as a complementary enabler of agile application development. Platforms such as Tulip and Mendix allow engineers, quality teams, and business users to design and deploy applications rapidly using visual modeling tools. These platforms support digital work instructions, operator performance monitoring, downtime tracking, and AI-driven defect detection, but face challenges related to scalability, validation, and compliance when deployed independently in regulated settings.

This paper proposes a hybrid framework that strategically integrates LCNC platforms with MES, positioning MES as the system of record for validated, GxP-critical processes and LCNC as a flexible innovation layer for rapid digitization. The framework draws on a qualitative, conceptual research methodology grounded in industry practice, architectural analysis, and regulatory frameworks, and is illustrated through comparative analysis and a proposed reference architecture. Organizations adopting this approach can reduce application backlogs, accelerate digital transformation, and maintain regulatory confidence, with the framework remaining extensible to other regulated manufacturing sectors.

In this study, a dry foam concrete mix was nano-modified, and based on it, foam concrete compositions were developed using lightweight porous aggregates—expanded clay gravel, foam glass, and polystyrene granules. It was experimentally determined that the introduction of silica sol in an amount of 0.001% relative to the cement mass resulted in an average increase of 15% in the compressive strength of the foam concrete. The developed foam concretes with porous aggregates are intended for thermal insulation and structural applications, with their average density formed in the range of 300 kg/m3 to 750 kg/m3. Research results showed that these foam concretes have a structural quality coefficient that is 30% or higher compared to foam concretes without aggregates, and they have the potential for effective application in the climatic and construction conditions of Uzbekistan.

Smart buildings are increasingly being promoted as a solution to rising global energy demand, yet many of them still suffer from inefficiencies in energy usage due to poor forecasting, suboptimal control strategies, and limited coordination between energy generation and consumption systems. At the same time, the integration of renewable energy sources such as solar, wind, and hybrid storage systems has introduced additional complexity because of their intermittent and unpredictable nature. In this context, artificial intelligence offers promising capabilities for improving energy optimization through demand prediction, adaptive control, and intelligent scheduling of energy resources. However, most existing studies focus either on AI-based energy management or renewable integration in isolation, with limited attention given to how these systems can be effectively incorporated into construction project management processes. This gap is particularly important during the design and planning phases, where early decisions significantly influence long-term building performance. This paper proposes a conceptual framework that integrates AI-driven energy optimization with renewable energy systems from a construction lifecycle perspective. The framework emphasizes data-driven decision support, lifecycle energy planning, and sustainability-aware project management. The key contribution lies in connecting energy modeling, AI techniques, and construction project decision-making into a unified approach aimed at improving both operational efficiency and environmental performance of smart buildings.

Dimensional stability is one of the key serviceability criteria for geopolymer concrete, especially when the binder is produced by a one-part route and cured at ambient temperature. This paper evaluates drying shrinkage, restrained shrinkage and compressive creep of one-part geopolymer concrete activated mainly with powdered sodium metasilicate anhydrous. The binder system was based on class F fly ash and ground granulated blast furnace slag (GGBFS), and the assessment focused on the effects of slag content, water-to-precursor ratio and solid activator composition. Five one-part mixtures were compared with an ordinary Portland cement concrete reference. Free shrinkage was monitored for one year together with mass loss and ultrasonic pulse velocity, while restrained shrinkage was measured on slab specimens by a photogrammetry-based procedure. Creep was evaluated under sustained compressive stress applied at 28 days. The slag-dominant mixture showed the highest one-year free shrinkage, reaching 1769 microstrain, whereas mixtures containing 60% fly ash and 40% GGBFS generally remained between 723 and 1091 microstrain. Lowering the water-to-precursor ratio from 0.45 to 0.35 reduced creep strain from about 1706 to 1551 microstrain and decreased the creep coefficient from 2.56 to 1.87. Restrained shrinkage of geopolymer mixtures was markedly lower than that of the Portland cement reference, with optimized mixtures remaining around 255-270 microstrain after one year. The results show that powdered sodium metasilicate can produce dimensionally stable one-part geopolymer concrete when slag content and water dosage are controlled to limit mesopore development, capillary pressure and early-age self-desiccation.

The fast-growing interconnection of digitized infrastructures and increasing sophistication of cyber-attacks have created unprecedented challenges for modern cybersecurity systems. Innovative technologies like Artificial Intelligence (AI), Digital Twins, and Federated Learning offer new avenues for developing smart, adaptive, and resilient cybersecurity solutions, which can deal with the emerging threats. However, conventional cybersecurity systems are based on the use of centralized architecture, reactive approaches to threat detection, and static security concepts. Moreover, contemporary security solutions relying on artificial intelligence experience various limitations concerning the protection of personal information, threat intelligence sharing among organizations, and creating realistic cyber-attacks for simulation. Therefore, there is an acute need for designing a scalable and secure cybersecurity ecosystem that can predict potential threats and help implement autonomous defensive measures. It is crucial to address the existing limitations and create a solution capable of increasing cybersecurity resilience. This paper presents the development of the Federated Digital Twin-Based Cybersecurity Ecosystem Using Generative AI for Predictive Attack Simulation and Defense. This framework incorporates digital twin technology in order to model the changing cyber-environment, utilizes federated learning to facilitate distributed threat intelligence sharing and employs Generative AI for effective attack simulation and defense generation.

Legacy healthcare information systems often operate through fragmented data models, monolithic applications, limited interoperability, and manually governed deployment practices that restrict scalability, security, and clinical responsiveness. This technical paper proposes an AWS-native modernization architecture that transforms legacy healthcare platforms into secure, interoperable, microservices-based systems supported by automated DevOps pipelines. The paper synthesizes research on electronic health record interoperability, FHIR-based integration, blockchain-enabled health data exchange, semantic data mapping, cloud-native systems, cybersecurity, healthcare analytics, and artificial intelligence to develop a structured modernization framework. The proposed model combines domain-oriented microservices, API-driven interoperability, container orchestration, automated CI/CD workflows, identity and access management, observability, and compliance-aware data governance. The analysis indicates that healthcare modernization should not be treated merely as infrastructure migration but as a socio-technical redesign of data, workflow, security, and operational governance. Findings suggest that AWS-native services can improve deployment reliability, horizontal scalability, auditability, and system resilience when combined with standardized clinical terminologies and secure interoperability layers. However, modernization also introduces risks related to vendor dependency, migration complexity, data quality, regulatory accountability, and organizational readiness. The paper concludes that a phased, security-by-design, interoperability-first architecture provides a practical pathway for healthcare institutions seeking to modernize legacy systems without disrupting clinical continuity.

This paper presents a separate material-design study focused on moisture balance, internal curing potential and resource-efficient raw-material selection for structural lightweight concrete intended for hot-dry climatic conditions. Unlike a conventional strength-centred comparison of fiber systems, the present work evaluates the technological role of expanded perlite, crushed ceramic brick waste, ceramic brick powder, limestone powder and silica fume in a low water-to-binder cementitious system. The study uses locally available mineral resources: CEM I 42.5N Portland cement, MK-85 silica fume, limestone powder, ceramic brick powder, quartz sand, crushed ceramic brick aggregate and pre-wetted expanded perlite. A moisture-balance approach was applied to estimate the potential internal water reservoir of porous components, and performance indices were calculated from the achieved properties of the optimized mixture. The designed concrete reached a slump of 16.5 cm, slump-flow diameter of 595 mm, average density of 1605 kg/m³, 28-day compressive strength of 57.0 MPa, flexural strength of 7.50 MPa, 28-day drying shrinkage of 0.475 mm/m and thermal conductivity of 0.475 W/(m·K). The calculated specific compressive strength was 35.5 MPa, while the strength-to-conductivity index reached 120 MPa·m·K/W. The results show that pre-wetted expanded perlite and ceramic waste components can be combined not only to reduce density and improve thermal efficiency, but also to stabilize the water balance of the cement matrix during early hydration in dry environments.