Institutional-Repository, University of Moratuwa
Welcome to the University of Moratuwa Digital Repository, which houses postgraduate theses and dissertations, research articles presented at conferences by faculties and departments, university-published journal articles and research publications authored by academic staff. This online repository stores, preserves and distributes the University's scholarly work. This service allows University members to share their research with a larger audience.
 Research Publications |  Thesis & Dissertation |  E- Books |
 UoM Journal Publications |  UoM Conference Proceedings |  Articles published in Scimago's Q1 journals |  UoM Research Reports |  Other Articles authored by UoM staff |
Recent Submissions
item: Conference-Full-text
AI chatbots for humanized service touch and customer satisfaction towards promoting AI-enabled banking interactions
(IEEE, 2025) Wikeshani, AHTJS; Jayamanna, RPAI; Wanasinghe, WDHBU; Arachchi, HADM; Samarasinghe, GD; Sisara, HT
This study examines the effects of AI chatbot communication techniques on customer satisfaction and intention to use AI enabled banking in the banking sector. Drawing on Expectancy Disconfirmation Theory (EDT) and Privacy Calculus Theory (PCT), the study investigates the effects of socially focused conversational styles and text-based communication mode on bank customers’ perceptions of humanness, happiness, and privacy concerns. A structured survey was completed by 135 Gen Z students, and Smart PLS software were used for analysis. The results demonstrate that both text-based and social-oriented communication styles significantly boost perceived humanness, which positively affects user satisfaction and intention to use. Additionally, it was demonstrated that privacy concerns moderated the relationship between pleasure and intention. The study fills in theoretical gaps and provides helpful information for chatbot designers and banking managers by applying the dual frameworks (EDT and PCT) in the context of AI-driven banking interactions. Enhancing user-centric chatbot capabilities while addressing privacy concerns may help to improve the customer experience and encourage broader AI use in digital banking.
item: Conference-Full-text
Predictive analytics for tea leaf aging and quality degradation
(IEEE, 2025) Bandara, RJ; Kuruppu, A
Tea quality directly influences flavor, marketability, and economic value. Traditional approaches to assessing tea leaf quality rely on manual inspection and cannot anticipate future degradation, leading to post-harvest losses. This paper introduces a two-stage, microservices-based predictive analytics system that empowers tea producers with forward-looking insights. In the first stage, a state-of-the-art object detection model processes harvested leaf images to classify quality into four tiers. In the second stage, a Random Forest classifier forecasts daily quality degradation over a fifteen-day horizon by combining leaf characteristics with environmental data-temperature, humidity, and rainfallfetched from public APIs and efficiently cached to minimize redundant calls. Explainable AI techniques distill each day’s prediction into the top three driving factors, presented in farmer-friendly language alongside actionable harvest recommendations. Deployed as serverless services and accessed through a mobile interface, the framework delivers scalable, low-latency predictions. This research addresses the critical need for proactive quality management in tea production by uniting image-based classification, environmental data integration, and explainability into a novel, end-to-end solution.
item: Conference-Full-text
ARGO-SLSA: software supply chain security in argo workflows
(IEEE, 2025) Mohomed, T; Ekanayake, I
Kubernetes has become the de facto standard when it comes to managing microservices. Automating complex, multi-step workflows is a common requirement in Kubernetes. Argo Workflows is a Kubernetes-native engine for managing these workflows in an automated fashion. These workflows generate artifacts such as executables, logs, container images, and packages. These artifacts require proper governance. Open-Source Security Foundation (OpenSSF), in collaboration with Google, introduced Supply-chain Levels for Software Artifacts (SLSA), a security governance framework that includes detailed technical requirements for producing artifacts. However, Argo Workflows doesn't have any built-in ways to provide the ability to incorporate the SLSA framework. This vacuum creates silos because practitioners need to rely on third-party tools to meet software supply chain security standards. This paper proposes a Kubernetes-native controller written to run in parallel to the existing open-source Argo Workflows to enhance the security of artifacts. Cryptographic signing and provenance attestations for the artifacts produced by the controller, which allows Argo Workflows to comply with SLSA standards. Evaluations were conducted in a real-world, self-hosted environment to demonstrate ARGO-SLSA’s ability to elevate artifacts to Level 2 of the SLSA compliance build track. Experimental results indicate that the ARGO-SLSA controller surpasses existing software supply chain security solutions.
item: Conference-Full-text
Exploring the application of modular product development in achieving multiple styling options in women's pants
(IEEE, 2025) Samarasekara, G; Seram, N; Mataraarachchi, R
The contemporary fashion industry is driven by a strong focus on sustainability and innovation. This research focuses on the Component-Sharing Modularity (CSM) technique as a transformative design approach for the women’s pants category. Rather than using a fixed single style garment, CSM introduces a system in which pant components are designed to be shared and reconfigured across multiple styles to achieve different style variations. This practice-based study uses prototyping to explore how CSM can offer a range of styling options for consumers using a limited number of components. By focusing on women’s pants, the project explores the modular technique in a familiar yet underexplored category, discovering possibilities for applying modularity in the apparel industry. The research highlights that CSM empowers designers to build collections with fewer garments and offers wearers the freedom to personalize their garments, using the multiple style options. This approach extends the effective use period of the garment, reducing consumption rates, encouraging responsible consumption and production. The challenges encountered when applying CSM to pants are addressed through experimentation, revealing new pathways for adapting modularity. The study presents CSM as both a technical and design solution that aligns with contemporary values of the fashion industry.
item: Conference-Full-text
Brachiation robot inspired by gibbon dynamics:insights for control across irregular bar spacing and heights
(IEEE, 2025) Perera, KDSHM; Amarasinghe, YWR; Jayathilaka, WADM; De Silva, E
Bio-inspired robotics draws inspiration from nature's solutions to locomotion, with brachiation robots replicating the swinging motion of primates to navigate discontinuous structures. While many systems often replicate the overarching motion, their control strategies often rely on complex full-state regulation rather than exploiting natural dynamics. This paper presents a simplified control approach where only the reaching hand is actively guided toward the next target while the body swings passively inspired by gibbon behavior. The method reduces control complexity while maintaining adaptability to irregular bar spacing and varying heights. The method is derived through dynamic modeling and validated in Gazebo simulations, with experimental verification on a physical prototype confirming feasibility. The results demonstrate that leveraging passive dynamics with minimal active control can achieve stable and efficient brachiation, offering a promising direction for energy-efficient robotic locomotion.