SLIIT International Conference On Engineering and Technology Vol. 04 [SICET] 2025
Permanent URI for this collectionhttps://rda.sliit.lk/handle/123456789/5064
Browse
Publication Open Access Design, Simulation, and Optimization of a Hybrid Hydrogen Fuel Cell–Battery Energy System for Sustainable Electric Vehicle Applications Using MATLAB Simulink(Faculty of Engineering, 2025-09-09) Samaranayake, W.A.K.L.The transition to sustainable transportation demands energy systems that are both efficient and environmentally friendly. This paper presents the design, simulation, and optimization of a hybrid energy system that integrates Proton Exchange Membrane (PEM) hydrogen fuel cells with lithium capacitor batteries for electric vehicle (EV) applications. The system aims to combine the high energy density of hydrogen with the fast response and recharge capabilities of advanced battery technologies to meet varying load demands efficiently. Using MATLAB Simulink, a hybrid model was developed to evaluate dynamic power sharing between the fuel cell and battery under variable driving conditions. A boost converter regulated the fuel cell output, while a bidirectional DC-DC converter managed power flow between the battery and the load. Maximum Power Point Tracking (MPPT) was implemented to optimize hydrogen fuel cell performance, enhancing energy efficiency under transient conditions (Dursun & Kilic, 2012). Simulation results demonstrated improved voltage stability, reduced stress on individual sources, and efficient energy utilization. Optimization of component sizing and control strategy further enhanced system response and fuel economy. These findings highlight the hybrid system’s potential for reducing EV range anxiety and promoting the use of renewable energy carriers such as hydrogen. This work contributes to ongoing research in sustainable mobility by offering a technically viable and scalable energy architecture that addresses the limitations of standalone battery and fuel cell systems. Future work will extend to hardware implementation and control refinement to accommodate real-world uncertainties and ensure robust performance.Publication Open Access A Comparative Study on TiO₂/Graphite–PEG and Graphite/Carbon Fibre- Paraffin Shape Stabilized Phase Change Materials for Thermal Energy Storage Applications(Faculty of Engineering, 2025-09-09) Dananjaya, V; Wen, Q; Abeykoon, PShape-stabilized phase change materials (SSPCMs) are promising candidates for latent heat thermal energy storage systems due to their high energy density and ability to prevent leakage during phase transitions. This study presents a comparative analysis of two SSPCM systems: TiO₂/graphite–polyethylene glycol (PEG) and graphite/carbon fibre/graphene–paraffin composites. Both composites were prepared by vacuum-assisted infiltration of molten PCMs into porous expanded graphite networks, with the addition of functional fillers to enhance structural integrity and thermal stability. Scanning electron microscopy (SEM) revealed distinct microstructural features for each system; TiO₂ nanoparticles were uniformly dispersed within the PEG matrix and anchored onto graphite surfaces, while carbon fibres and graphene nanoplatelets formed a hierarchical interconnected network within the paraffin-based composites. Differential scanning calorimetry (DSC) demonstrated that both systems preserved high latent heat storage capacities with slight shifts in phase transition temperatures compared to pure PCMs. Thermogravimetric analysis (TGA) showed improved thermal stability of the SSPCMs relative to neat PCMs, with filler composition significantly affecting degradation onset temperatures. In TiO₂/graphite–PEG composites, DSC analysis showed melting temperatures of 61.4-62.7 °C and solidification temperatures of 53.1-54.0 °C, with latent heats of 185-210 J g⁻¹ depending on TiO₂ content. Graphite/carbon fibre/graphene–paraffin composites exhibited melting temperatures of 54.8-55.6 °C and solidification temperatures of 48.9-49.7 °C, with latent heats of 140-160 J g⁻¹. Thermogravimetric analysis revealed improved degradation onset temperatures: TiO₂/graphite-PEG composites showed higher thermal stability compared to pure PEG, while carbon fibre/graphene–paraffin composites exhibited enhanced thermal resistance relative to pure paraffin. The TiO₂/graphite-PEG composites exhibited higher latent heat capacities and enhanced thermal resistance, whereas the graphite/carbon fibre/graphene–paraffin composites provided superior mechanical reinforcement and phase change reliability. These findings offer insight into the design optimization of SSPCMs tailored for specific thermal management applications.Publication Open Access Development Of An Ai-Based Model With Low Computational Complexity For Accurate Solar Energy Forecasting(Faculty of Engineering, 2025-09-09) Chandrasinghe, S; Fernando, NThis paper introduces a short-term solar energy forecasting model that is designed with a focus on low computational complexity and addresses the challenges posed by fluctuations in solar energy generation, which are significantly influenced by environmental factors. These fluctuations can lead to instability when solar power generation systems are integrated into national energy grids, creating difficulties in maintaining a balanced supply and demand. If solar energy generation can be accurately forecasted before fluctuations occur, potential issues can be identified in advance, allowing for better management of the energy system, including optimizing storage facilities when energy generation is high. Current solar energy forecasting systems face significant challenges due to their high computational complexity, which results in increased power consumption and lower accuracy. To address these issues, this study focuses on the development of an artificial intelligence (AI)-based forecasting model using an Artificial Neural Network (ANN). The goal is to reduce the computational complexity of the model while maintaining high accuracy. To achieve this, various data analysis and complexity reduction techniques, such as variable reduction, pruning, and quantization, were applied. The performance of the optimized AI model was evaluated by comparing the forecasted values to actual solar energy generation data. The results demonstrate that the proposed model successfully reduces computational complexity while maintaining a satisfactory level of accuracy. This optimization makes the model more suitable for real-time forecasting, particularly in resource-constrained environments, and provides a more efficient approach to solar energy management. The findings of this study suggest that AI-based forecasting models can play a critical role in enhancing the integration of solar energy into national grids, ensuring a more reliable and sustainable energy supply. Further research could explore additional optimization techniques and the introduction of generalization techniques to improve transferability of the model and applicability across diverse geographical regions. Additionally, focus on utilizing AI techniques that minimize computational complexity without compromising the accuracy of the model, aiming to maintain high forecasting precision while optimizing the efficiency of the system.Publication Open Access Enhancing Modern Education Through an AI-Integrated Learning Management and Support System (LMSS)(Faculty of Engineering, 2025-09-09) Rashminda, JThe rapid advancement of educational technologies has underscored the limitations of conventional Learning Management Systems (LMS) in effectively supporting the evolving demands of learners and educators. While traditional LMS platforms primarily focus on content delivery and administrative tasks, they often lack the capacity to foster active engagement, facilitate meaningful collaboration, and promote participation in broader learning experiences. This paper presents the design and functional implementation of a prototype for a Learning Management and Support System (LMSS), an AI-enhanced platform built to address these limitations by offering a more holistic and student-centred approach to digital education. LMSS integrates course management with interactive features that encourage student collaboration, peer-to-peer communication, and involvement in academic and extracurricular events. These capabilities are designed to support a more engaging and socially connected learning experience while also simplifying instructional workflows for educators. The system incorporates adaptive learning tools and real-time insights to better align learning processes with individual needs and institutional goals. This paper reviews the existing literature, highlights gaps in current LMS implementations, and details the development methodology, architecture, and feature set of LMSS. The system’s anticipated impact is grounded in established research findings demonstrating that adaptive learning approaches can significantly enhance student engagement, AI-driven early intervention can improve retention rates among at-risk learners, and realtime analytics can reduce instructor workload related to feedback provision. By integrating these evidence-based practices into a unified platform, LMSS is designed to foster learner motivation, deepen engagement, and support teaching effectiveness. Ethical considerations such as user privacy and data governance are also addressed to ensure responsible and transparent use.Publication Open Access Keynote 01:Optimization Methods in Computational Intelligence for Joint Wireless Channel Parameter Estimation over Advanced Radio Interface Technologies(Faculty of Engineering, 2025-09-09) Affes,SofieneAdvanced Radio Interface Technologies (RITs) combine broadband signalling—hence multicarrier operation and richly multipath propagation—with multi-antenna transceivers. In these regimes, joint estimation of channel parameters (angles of arrival/departure, delays, Doppler/frequency offsets, gains/phases, etc.) becomes a central yet challenging inference problem. Objective or cost functions are often nonconvex, multimodal, and simulator-defined, with scarce gradients, tight pilot budgets, and low signal-to-noise ratios (SNRs). Therefore, among computational intelligence (CI) categories that encompass 1) neural networks and 2) fuzzy systems, the third or 3) population-based and bioinspired optimization (PBO/BO) methods – such as particle swarm optimization (PSO), differential evolution (DE), genetic algorithms (GA), grey wolf optimizer (GWO), and related swarms – have gained traction as global search engines that either directly minimize maximum-likelihood (ML) or mean-square error (MSE) criteria or act as robust initializers for hybrid pipelines. In this talk, first we integrate a disciplinary taxonomy relating artificial intelligence (AI), optimization, and Monte Carlo inference to place CI and PBO/BO within a broader computational context worth contemplating. Then we survey the current state of the art on CI optimization for wireless channel parameter estimation and analyze the strengths and weaknesses of each CI subcategory versus the others and against conventional estimation methods. We synthesize algorithmic patterns, objectives, accuracy, convergence/complexity trends, and empirical findings, etc., over advanced RITs, and we discuss most recent progress and open challenges.Publication Open Access Development of an AI-Based Model with Low Computational Complexity for Accurate Wind Energy Forecasting(Faculty of Engineering, 2025-09-09) Dilshan, S; Fernando, NMost countries primarily relay on fossil fuel for electricity generation, leading to fossil fuel depletion and environmental pollution. The countries are developed technologies for renewable energy generation. The wind energy being promoted as a superior renewable energy. However, wind energy has its challengers, particularly uncertainty that can affect overall system stability. The accurate short-term forecasting of wind energy was crucial for ensuring grid stability. Both physical and AI-based models can effectively be utilized for wind energy prediction. AI-based methodologies have shown superior effectiveness, efficiency, and accuracy when compared to traditional physical models. The lightweight AI-based forecasting model was particularly significant for processing devices, enabling faster computations and substantially more cost-effective forecasting. The research utilized simulation software to develop an Artificial Neural Network (ANN) model, initially incorporating eight meteorological parameters. Four of these parameters showed weak correlations and were subsequently removed from the model. Further optimization was achieved through pruning and quantization techniques, significantly reducing computational complexity. The optimized model demonstrates a notable reduction in both training time by 92.69% and inference time by 63.83%, while maintaining accuracy with only a marginal decrease of 3.99% compared to the initial model. These improvements were achieved with minimal loss in predictive accuracy, significantly reducing computational complexity. The study concludes that the optimized ANN model is wellsuited for real-time wind power forecasting, offering a balance between accuracy and computational efficiency. This approach not only facilitates better grid management but also extends the applicability of AI-based forecasting to devices with limited processing capabilities. Future work could explore additional complexity reduction techniques and broader deployment scenarios.Publication Open Access Towards Safer Elderly Care: A Convolutional Neural Network Solution for Fall Detection(Faculty of Engineering, 2025-09-09) Kalupahana R.W; Maduranga M.W.PAs modern life becomes increasingly busy, computer vision-based monitoring systems have become essential, particularly in elderly care. This paper presents the development of a robust fall detection system using deep learning techniques, specifically a convolutional neural network (CNN) that processes RGB images to accurately distinguish between fall and non-fall events. The model is trained and validated on a dataset categorized into two classes: fall and non-fall. By utilizing convolutional and pooling layers, CNN effectively learns hierarchical representations of the input data, capturing both low-level and high-level features crucial for accurate fall detection. The key stages of this approach include data acquisition, pre-processing, and model training. The model's performance is evaluated using precision, recall, and F1-score metrics, demonstrating high accuracy, which is further enhanced through data augmentation, pre-processing, and crossvalidation techniques. A confusion matrix analysis confirms the model's effectiveness in correctly classifying instances across both classes. The system also extends its capabilities to video analysis by extracting frames at 30-second intervals, ensuring continuous and comprehensive monitoring. This research highlights the potential of deep learning to enhance safety and care for the elderly, offering a reliable solution for real-time fall detection. The findings underscore the importance of integrating advanced technologies into healthcare, paving the way for future innovations in monitoring and assistance systems.Publication Open Access Piezoelectric Energy Harvesting in Interactive Environments: A Case Study Using a Playpen-Based Prototype(Faculty of Engineering, 2025-09-09) Bandara, H. M. U. J.; Fernando, M. N. V.; Samaranayake, H. D. K.As modern cities evolve towards sustainable and efficient infrastructure, implementation of micro-energy harnessing technologies caught the attention of researchers. This is mostly because non-renewable energy sources are being depleted, and the consequences of them are also marginal to the world. Piezoelectric materials are known for their conversion ability of mechanical stress, such as motion, vibrations, pressure, to electrical energy. This study provides a piezoelectric mechanism-based prototype that demonstrates the suitability of converting mechanical energy to electricity, especially footsteps from people, while being low maintenance and cleaner. In this design, 70 PZT piezoelectric disks are implemented beneath a playpen floor, suspended by a spring mechanism and further suspension components. Output voltages of 9.2V to 11.4V could be observed from this demonstration, with the help of 2-5kg loads applied. 76 LEDs were used to visualize this power in real-time, which also offers interactive feedback for children. The output analysis shows a low-current and somewhat non-linear power output. These results validate the possible outcome of deploying a piezoelectric system in high footfall infrastructure and recreational spaces. Even though it is low powered, such systems can enhance built environment functionality, and further improvements in the design can increase the output of such designs. This project represents a small-scale demonstration of this unpopular yet effective technology.Publication Open Access Multifactorial Drivers of Chronic Kidney Disease of Unknown Etiology (CKDu): A Review of Cadmium Exposure, Ultraviolet B B Radiation, and the Potential Role of Vitamin D Toxicity(Faculty of Engineering, 2025-09-09) Ariyawansha, R.T.K.; Basnayake, B.F.A.; Dharmasena, D.A.N.; Siribaddana, SThere are many research studies conducted to determine the cause/origin of CKDu for preventing this increasingly occurring disease, particularly among poor farmers. The disease is known to be multifactorial with heat stress enduring period > 3 months, but there is difficulty in distinguishing between harsher environments than endemic locations. This study identified that the likely cause is UVB actuating Vitamin D Toxicity (VDT), thus affecting the kidneys. Hence, a focused literature review was undertaken to find the links between cadmium (Cd), calcium, fluoride, enzymes, inhibitions, and the like. Moreover, 2nd law of thermodynamics was applied to determine the entropy differences between cold and hot source. The mean values of climate models were obtained from one publication on radiative forcing (RF) in the tropopause of 1.28 Wm- 2 and climate feedback (CF) 0.25 Wm-2K-1. The energy of RF was used to determine entropy value 𝑆𝑅𝐹(𝑈𝑉𝐵)at 𝑇2 = 230𝐾 as mean atmospheric temperature and the maximum temperature, 𝑇1 at locations. It was then equated to the energy value of UVB, 𝑄𝐶𝐹(𝑈𝑉𝐵) to be found between 1/273 and 1/𝑇1. It was also validated using CF. The endemic location resulted 3.697 Wm-2 at 303K, and low RH compared to 3.239 Wm-2 at 311 K high RH. Although there is much comfort in endemic location, the chances of VDT or heat stress are higher more so with Cd inhibition of enzyme 7-dehydrocholesterol reductase (DHCR7), which is crucial for cholesterol synthesis. Instead, 7-dehydrocholesterol in excess switches more to form VDT, causing symptomatic hypercalcemia. Cadmium can disrupt vitamin D metabolism, contributing to osteomalacia and osteoporosis, actuating hypercalciuria, an indirect marker of low-level cadmium exposure. The kidneys, already compromised due to cadmium (Cd) accumulation and reabsorption during systemic distribution, ultimately eliminate Cd via the urine. Notably, no significant Cd accumulation is observed in end-stage renal tissues. Further basic research is required to elucidate the VDT in response to UVB exposure.Publication Open Access A Data-Driven Approach to Predicting Ischemic Heart Disease Risk in Monaragala: Integrating Lifestyle and Symptom Factors with Machine Learning(Faculty of Engineering, 2025-09-09) Meddepola, M.A.R.L.; Wickramasinghe, B.M.G.S.T.S.K.Ischemic Heart Disease (IHD) remains a leading cause of mortality worldwide and presents a critical challenge in underserved rural areas such as Monaragala, Sri Lanka. Traditional IHD prediction methods predominantly depend on clinical diagnostics like ECGs and blood tests, which are often unavailable or inaccessible in such regions. This study aims to bridge this gap by developing a machine learning-based prediction model that utilizes only lifestyle and symptom-related data, eliminating the need for invasive clinical procedures. A dataset comprising lifestyle habits (e.g., diet, smoking, alcohol use, exercise) and symptom indicators (e.g., chest pain, fatigue, dizziness) was collected via surveys. Feature selection using Logistic Regression identified the top eight most relevant predictors. Five machine learning algorithms, Logistic Regression, K-Nearest Neighbors, Support Vector Machine, Decision Tree, and Random Forest, were trained and evaluated. Among them, the Random Forest model achieved the highest performance with an accuracy of 83.5%, precision of 0.86, recall of 0.78, and F1- score of 0.81, demonstrating strong predictive capability based solely on non-clinical features. In addition, a web-based self-assessment tool was developed to make the model accessible to the public, particularly targeting individuals in rural areas with limited healthcare access. The tool enables users to input basic lifestyle and symptom information and receive a real-time risk assessment. The findings confirm that the model leveraging lifestyle and symptom data can effectively identify individuals at risk of IHD. This approach supports the development of scalable, low-cost, and user-friendly screening tools that can enhance early detection and preventive care, especially in rural and resource-constrained settings.Publication Open Access Li-ion Battery Cooling - A Computational Study of Different Phase Change Material Configurations(Faculty of Engineering, 2025-09-09) Adikaram, S; Nasser, A; Vallés, C; Abeykoon, COverheating of Li-ion batteries in Electric Vehicles (EVs) degrades performance and reduces lifespan. Hence, energyefficient and reliable Battery Thermal Management Systems (BTMS) are required. This paper investigates the use of Phase Change Materials (PCMs), a passive cooling method with high heat storage capacity, for the thermal management of prismatic Li-ion battery cells in EVs. This computational study models the influence of buoyancy-driven convective flow on the PCM cooling performance, compared against thermal conduction-only models. In addition, this study investigates how convective flow influences the cooling performance with variations in cell orientation between vertical and horizontal alignments. n- Octadecane is used as the PCM, and Computational Fluid Dynamics (CFD) simulations were conducted with the Solidification and Melting model in ANSYS Fluent. A 12 mm PCM layer placed around the cell periphery reduced the centre temperature after 1800 s by 2.7 K in the vertical orientation and 3.7 K in the horizontal orientation compared to air-cooling. The effect of natural convection was more pronounced in the horizontal orientation, providing superior cooling performance relative to the vertical case. When the same PCM volume was used to fully enclose the cell, the cooling effect was further enhanced, achieving a maximum temperature reduction of 8.3 K within the first 1800 s. The findings demonstrate that natural convection significantly enhances the PCM-based cooling effectiveness, particularly in horizontally oriented cells, while thinner PCM layers with increased heat transfer area promote faster melting and improved cooling performance.Publication Open Access Diabetic Retinopathy Screening Using Image Processing(Faculty of Engineering, 2025-09-09) Wijesekara, W.M.M.P.; Karunarathna, A.E.S.H.; Haluwana, H.M.R.M.K.; Jayawardhana, S.M.M.SDiabetic retinopathy, a grave consequence of diabetes mellitus, has emerged as the leading cause of visual impairment worldwide. This ocular condition arises from the deterioration of blood vessels situated behind the retina and progresses insidiously, ultimately leading to blindness. Early detection is paramount in mitigating vision loss among afflicted individuals. In this study, we propose three distinct approaches Support Vector Machines (SVM), Convolutional Neural Networks (CNN), and Residual Networks (ResNets) for the accurate detection of diabetic retinopathy. Our aim is to determine the most effective model for this purpose, thereby improving screening efficiency. Utilizing a pre-processed dataset sourced from Kaggle, we conducted comprehensive experiments to evaluate the performance of each model. This curated dataset was instrumental in optimizing the classification algorithms. Our findings reveal notable disparities in the performance of these models. Through meticulous testing and validation, we sought to identify the model exhibiting the highest accuracy in diabetic retinopathy detection. Leveraging a dataset comprising 2750 retinal images, our experiments yielded accuracy values of 68% for SVM, 74% for CNN, and 63% for ResNet.Publication Open Access Evaluation of the Knowledge Base in Agriculture and Food to Reduce and Prevent Chronic Kidney Disease of Unknown Etiology (CKDu)(Faculty of Engineering, 2025-09-09) Ariyawansha, R.T.K.; Basnayake, B.F.A.; Dharmasena, D.A.N.; Gamage, AKidney disease is a growing global problem, more so in tropical regions. The cause of CKDu is multifactorial and influenced by heavy metal (HM) contamination, inhibiting essential enzymatic reactions. Fertilizers and water contamination are believed to cause the disease. This study aimed to review the existing knowledge base, focusing on a transitional approach to advanced technologies with the least HMs and to use justifiable scientific reasoning supported by published data, to used to demonstrate the movement of Cadmium (Cd) at both low and high concentrations from applied fertilizer through the soil to grain and rice. The quantity of fertilizer applied per ha with the given Cd levels was equated to Cd concentrations in the harvested grain and rice per ha, considering positive or negative contributions from the soil. Weekly consumption levels of rice at the threshold limits by an average Sri Lankan were determined for low and high Cd levels in rice using the tolerance limits of two international standards. It is best to characterize watersheds and determine the movement of nutrients and HM in ferruginous soils. Hinderance to phosphate immobility in these soils can be overcome by applying biochar biofertilizer with possible enrichment of biofilm biofertilizers to replace totally inorganic fertilizers contaminated with HMs. Cd levels of 836.25 and 393.75 of the two publications equate to the assumed harvest: lowest 21.22, average 385.13, and the highest 1246.10 mg Cd ha-1. Allowable standards indicate that the weekly limit of a Sri Lankan to consume rice is 300 g, containing a high concentration of 0.2618 mg Cd kg-1 and 1kg or 604 g, having 0.1339 mg Cd kg-1 for an average harvest of 4350 kg.ha-1. Water contains HM, particularly arsenic from fertilizer and pesticides. Recommended researching while implementing phytoremediation, mechanized farming, preventing UVB, Integrated Pest Management (IPM), and organic agriculture with supporting technologies of watershed resource management.Publication Open Access Development of Low-cost Slipper by using NR/EVA Blend with Recycling Materials for reducing Environment Pollution in Footwear Industry(Faculty of Engineering, 2025-09-09) Randika K.G.; Perera K.P.M.; Gunaratne R.D.This study aims to develop a low-cost slipper compound by blending low grade Natural Rubber (NR), Ethylene Vinyl Acetate (EVA), and recycled LDPE plastics granules with crumb rubber in different phr (parts per hundred rubber) amounts. Low grade natural rubber (off grade brown scrape) and ethylene vinyl acetate (19 wt.% of vinyl acetate) were used during formulation in order to reduce cost. During this compounding process, polymeric material and other chemical ingredients were masticated by using a kneader and two roll mills then sheet was prepared by using calendaring techniques, eventually curing was performed by using a compressing molding method. Blowing agents were used to obtain the Slipper sheets’ inter cellular structure. Peroxide curing system has used due to natural rubber blend with ethylene vinyl acetate. When preparing compound batch, different phr amount of crumb rubber and recycled LDPE plastics granules blended. Firstly, crumb rubber sheets which were punctured and waste scrap sheets were obtained, then converted into 30 mesh size small particles by using grinding and crush method. Hardness and Abrasion tested of prepared slipper. After curing process higher hardness value observed when increasing crumb rubber phr. As particle size increases, there was a tendency of asymmetrical spread of compounding ingredients through the mixture and this was mitigated by additions of processing oils which increased the dispersion of the particles. In summary, through this work, ideal compounding formulation with phr values was able to determine that can be used to manufacture to low-cost slipper sheet at industrial scale.Publication Open Access Predicting Cognitive Test Performance from New Onset Behavioral and Personality Changes in Adults over 50 using Post-Selection Boosted Random Forest Classifier(Faculty of Engineering, 2025-09-09) Mervyn M.; Welhenge A.; Creese B.Mild Behavioral Impairment (MBI) refers to neuropsychiatric symptoms of various severity levels that might not be discovered by conventional psychiatric nosology. These symptoms should persist for more than six (06) months. MBI is typically observed in adults of age 50 and above. This study investigates the prediction of cognitive test performance of cognitive and behavioral changes in adults over 50 years of age using a post-selection boosted Random Forest (RF) Classifier. The baseline cognitive aging data of the Simple Reaction Time (SRT) metric and Mild Behavioral Impairment Checklist (MBI-C) from the ongoing PROTECT study in the United Kingdom was used to classify the participants’ cognitive ability into five classes. Using the post-selected boosted RF classifier, the study obtained an accuracy of 96.26% which was an improvement compared to the 95.52% accuracy obtained by the RF classifier. These findings suggest that machine learning-based prediction models can provide valuable insights into analyzing the cognitive decline of adults of a late age.Publication Open Access Predictive Modeling for Personalized Cancer Therapy Using Reinforcement Learning(Faculty of Engineering, 2025-09-09) Edirisinghe M.M; Gunarathne, J H M S MAdaptive therapy is transforming cancer treatment by enabling dynamic, patient-specific interventions that adapt to tumor progression and individual variability. Unlike traditional fixed-dose regimens, adaptive therapy leverages the evolutionary dynamics of tumors to extend treatment effectiveness and delay resistance. Reinforcement Learning (RL), an area of artificial intelligence focused on sequential decision-making, offers a robust framework for optimizing these adaptive strategies. RL can learn optimal treatment policies by interacting with computational models of tumor growth and drug response, continuously adjusting regimens based on observed tumor states, resistant cell populations, and biomarkers. This approach allows for the creation of personalized therapies that maintain long-term tumor control while minimizing toxicity and the emergence of resistance. The integration of RL into predictive modeling for cancer therapy represents a paradigm shift, enabling smarter, safer, and more effective treatments that are dynamically tailored to each patient’s evolving disease. This paper reviews the foundational concepts of adaptive therapy and RL discusses tumor modeling approaches, examines RL algorithms, and addresses current challenges and future directions in the field.Publication Open Access The Influence Of Project Managers' Decision-Making Styles On Schedule Variance In Building Construction Projects(Faculty of Engineering, 2025-09-09) Pulasinghe, K; De Silva, PThis research focus to assess the relationship between project managers decision making styles and schedule variance in building construction projects in Sri Lanka. Timely completion of construction projects is one of the major performance indicators, yet delays are a long-standing issue in Sri Lankan construction projects. Though there are many causal factors, decision-making styles of project managers have not been studied yet, creating a significant knowledge gap with regard to their influence on project schedule variance. This study attempts to analyze the relationship between project managers' decision-making styles: directive, analytical, conceptual, and behavioural and schedule variance in Sri Lankan building construction projects. A mixed-methods research approach was adopted. Primary data were gathered by holding semi-structured interviews with nine industry practitioners and a questionnaire survey of 50 respondents covering key project roles. To analyze the data code based content analysis and descriptive statistical tools (percentage count, mean, weighted average etc.) were used and to examine the relationship between decision-making styles and schedule variance Pearson correlation was conducted. Results revealed that decision making styles play a significant role in influencing project schedules. However, it was found that directive and behavioural styles are most prevalent and successful styles in the Sri Lankan context. The data revealed both positive and negative influences of managerial decisionmaking styles on schedule performance. These results contribute to the link between leadership and project performance and make a theoretical and practical contribution by revealing decision-making as a key influence to minimize schedule variance.Publication Open Access Development of an AI-Based Model with Low Computational Complexity for Accurate Load Demand Forecasting(Faculty of Engineering, 2025-09-09) Hettiarachchi, D.R.A.; Fiernando, NThis research addresses the challenge of short-term load demand forecasting in microgrids, where renewable energy unpredictability destabilizes power systems. Current forecasting models often suffer from high computational complexity, resulting in increased power consumption and reduced real-time applicability. To overcome these limitations, this study develops and optimizes an Artificial Neural Network (ANN)-based shortterm forecasting model with significantly reduced computational demands. In this study, a model was constructed utilizing historical operational data from a microgrid system. To optimize the computational efficiency of the model, various techniques were applied to reduce its complexity. The model’s performance was systematically evaluated using appropriate performance metrics. The experimental results demonstrate that the proposed approach significantly decreases the computational complexity of the final model, while preserving an acceptable level of accuracy when compared to the original, unoptimized model. The practical implications of this research include enabling real-time demand forecasting on resource-constrained microgrid controllers and edge devices, facilitating more efficient energy management in sustainable power systems. Future work will focus on enhancing the model's generalization capabilities by incorporating additional geographical and climatic factors, enabling accurate demand forecasting across diverse microgrid environments beyond the specific conditions of the initial dataset.Publication Open Access FreshSight: An Accessibility-Focused Approach to Produce Freshness and Shelf- Life Detection for Food Safety and Waste Reduction(Faculty of Engineering, 2025-09-09) Fernando, W.P. R.; Kirupananda, A.Colour-blind individuals encounter daily challenges, particularly in distinguishing colour-based indicators of food spoilage. This limitation significantly impacts their ability to assess the freshness and safety of fruits and vegetables accurately. Concurrently, global concerns regarding food spoilage have intensified, with millions worldwide affected by foodborne illnesses each year. The modern lifestyle, characterized by its rapid pace and time constraints, exacerbates this issue, often leading to unnoticed spoilage and substantial waste. The resulting annual waste, estimated at one-third of all edible food, imposes significant societal and environmental burdens, underscoring the urgency for effective solutions. FreshSight enables users, including those with colour vision deficiencies, to assess the condition of fruits and vegetables through a Convolutional Neural Network (CNN)- based real-time image analysis engine and an intuitive interface. This system provides immediate visual feedback to help users make informed decisions and avoid the consumption of spoiled produce. It also offers inclusive design features that support individuals with visual impairments. Beyond individual benefits, FreshSight promotes responsible food handling and contributes to the broader goal of sustainable food systems. By combining advanced technology with user-centered design, the solution enhances both safety and accessibility in everyday food-related decisions. In addressing the critical challenges of food safety, inclusivity, and waste reduction, FreshSight aims to support healthier lifestyles and contribute positively to environmental and societal well-being in the modern world.Publication Open Access Penetratingz the Defenses: An Investigation into the Achilles' Heel of HTTP and SSH(Faculty of Engineering, 2025-09-09) Joshi, A.P; Bandrevu, S; Kaur, N; Sharma, I; Maduranga, M.W.P.; Wanniarachchi, W A A MAs the yber threat increases, it becomes necessary for organizations to start securing their digital valuables and infrastructures. This research is mainly about analysing the weaknesses lying in the vulnerabilities of the HTTP and SSH protocols. Investigation here is into how the intruder can escalate his privileges and illegally access computers. Under open-source tools like Netcat and Gobuster, the article examines the vulnerability-assessment methodologies culminating in root access to the target machine. This paper emphasizes the need for proactive security measures and gives recommendations on improving defences against future attacks. The study, as bright as the findings may be, awaits empirical dimensions to affirm the proposed measures.