Journals of Universiti Tun Hussein Onn Malaysia (UTHM)
Not a member yet
6081 research outputs found
Sort by
Burst-Aware Weighted Fair Queueing for Serverless Inference: Mitigating Noisy Neighbor Effects in Multi-Tenant Systems
Full text linkMulti-tenant serverless inference often devolves into noisy-neighbor scenarios where a single tenant’s bursty LLM batch floods the fleet, pushing interactive calls beyond their latency budgets. We are proposing a Burst-Aware Weighted Fair Queueing (BWFQ) - a scheduler that requires only two counters per tenant (tokens earned, tokens spent) and a constant-time heap pop to pick the next invocation. In BWFQ, we use the classic token-bucket shaper where tokens accumulate at a tenant-specific base rate and are reduced on each dispatch. When a tenant exhausts all its tokens, its requests are queued, giving chances to other quieter tenant s to run. Techniques described in other papers like Dominant-Resource Fairness, BWFQ requires neither per-invocation resource profiling nor multi-dimensional share accounting, making it easy to integrate onto existing Lambda-style dispatchers. To evaluate our algorithm, we built a prototype using AWS Lambda and observed that BWFQ reduces the P99 latency gap between interactive and batch tenants from 8.5s to 2.1s; a 4.0X improvement, while preserving 94% of the throughput achieved by First-Come-First-Serve. The algorithm adds only 35 µs of scheduling overhead per decision and fits in approximately in 150 lines of Go code. These results demonstrate that a simple token-bucket fair queueing is a practical, immediately usable step towards building fairness in production serverless inference
Synthesis and Characterization of Al-SiO2 Composites
Full text linkThe paper presents studies aimed at producing aluminum matrix composites reinforced with amorphous silica particles. The possibility of producing composites of the Al-SiO2 (5 wt. %) system using intensive mixing casting and semi-solid metal casting methods has been established. Semi-solid metal casting followed by squeeze casting demonstrated the highest efficiency. The potential of using magnesium as a surface-active additive to remove oxygen from the surface of the dispersed particles and improve the mechanical properties of the composite material during heat treatment has been demonstrated. It has been found that the resulting composites exhibit a uniform distribution of dispersed silica particles throughout the metal volume. Furthermore, they possess hardness, corrosion resistance, and specific gravity values that exceed those of the original aluminum alloy. Composites produced using the developed technique may find application in various transportation engineering fields, as well as in aviation and aerospace industries
The Evolution of Metaheuristic Research: A Bibliometrics Analysis of Research Trends in Computer Sciences
Full text linkThe purpose of this paper is to examine the development of research trends and to propose potential future paths by conducting a bibliometric analysis of metaheuristic studies. Enhancing the sector\u27s importance and giving readers a clear understanding of its development. The research uses a bibliometric approach, collecting its information from the Scopus database. Articles published in academic journals on metaheuristic research between the years 2015 and 2023 will be analyzed for this study. Data is collected, analyzed, and interpreted so that conclusions can be drawn. Important insights into the development of metaheuristic studies emerge from the analysis. Global optimization, heuristic approaches, scheduling, genetic algorithms, evolutionary algorithms, and benchmarking are only a few of the overarching research issues that this study reveals. There is a growing interest in new fields of study, such as adaptive neuro-fuzzy inference, forecasting, feature selection, biomimetic, and the exploration-exploitation trade-off. The results shed light on where the field of metaheuristics is at the moment and where it needs to go in the future. The study\u27s dependence on Scopus data and its refusal to include data from other sources are two of its acknowledged drawbacks. It draws attention to the need for more study to address these limitations and recommends possible directions for future studies, such as adding different data and looking at different applications. This research appends to the frame of knowledge by providing a thorough bibliometric examination of how metaheuristic studies have developed. It emphasizes the value of metaheuristic algorithms as optimization instruments and their influence on resolving complex issues. Researchers, practitioners, and policymakers can all benefit from these findings, as they shed light on current and future paths in metaheuristics research
Urban Mobility, Tourism, and Liveability in Penang: Anticipated Benefits of the Proposed Light Rail Transit (LRT) System
Full text linkPenang Island, Malaysia, continues to grapple with persistent urban challenges, including severe traffic congestion, constrained public transport accessibility, and growing environmental concerns. In response, the Penang state government has proposed a Light Rail Transit (LRT) system under the Penang Transport Master Plan (PTMP) to address these critical issues. This conceptual paper examines the potential urban benefits of the proposed LRT project, with a particular emphasis on its capacity to reduce traffic congestion, strengthen the tourism sector, and enhance urban liveability. Through the analysis of secondary data, regional infrastructure trends, and comparative insights from established LRT systems in Southeast Asian cities, the study highlights the anticipated socio-economic, environmental, and mobility-related impacts of the project. The findings suggest that, if effectively planned and implemented, the Penang LRT system could play a transformative role in fostering a more sustainable, connected, and inclusive urban environment, contributing positively to the island’s long-term economic growth and quality of life
An Analysis of Logistic Cost Structure in Corn Commodity Supply Chain (Zea Mays L) in East Java Using Activity Based Costing (ABC) Method
Full text linkEast Java is Indonesia\u27s biggest corn producer in 2022, with a percentage of 26,24. The El Niño phenomenon of 2023 caused a decrease in corn production in Indonesia. El Nino phenomenon in non-farming land can add logistic costs, especially in procurement. The goal of the logistic cost structure analysis is to determine the proportion of logistic costs. This research was conducted in three regencies in East Java, a representative of the corn producer centre. A total of 98 respondents were included in the study. The research sample was selected based on convenience sampling and snowball sampling. The Activity Based Costing (ABC) method calculates the logistic-cost structure. The structural analysis of logistic costs used descriptive statistics. The analysis of the logistic cost structure showed that the most significant proportion of logistic cost was procurement activity, with a percentage of 63,09%. The activity at the farmers\u27 tier with the most significant cost proportion was procurement activity, with a percentage of 69,73%. Transportation was the activity with the most considerable logistic cost proportion in the collector trader and large trader tiers, with proportions amounting to 52,31% and 83,01%. The right supply chain strategy for corn commodities in East Java was push-based
Function of Construction Waste as A Replacement Material for Stones in Stone Column
Full text linkSustainability is an important aspect of the Civil engineering domain and mainly focuses on the reduction of carbon footprints. One of the major contributors to carbon footprint is construction and demolition (C&D) waste, which increases with urbanisation. With an increase in awareness of recycling, the usage of C&D waste is used as a replacement material for stones in a stone column. The stone column is a conventional ground improvement technique used to stabilise the soil by improving its strength and settlement behaviour. The C&D waste of M30-graded crushed concrete and the brick is considered and used as a replacement material to make columns of two consistencies (0.4 and 0.6). The other waste material of C&D waste is removed and crushed to a size of 5 mm and used as a column material. A load-penetration study is conducted in California Bearing ratio (CBR) mould by developing a column in the center of the mould with 100% of conventional aggregate, crushed concrete and bricks which is compared with the virgin clay. Due to the interlocking behaviour of the column materials, the initial behaviour of the load-penetration behaviour follows the same pattern. With an increase in load intensity, the load bearing of crushed concrete is similar under higher consistency. As the water retention of brick is more for lower consistency, the load bored by 100% crushed brick is the same as that of stone aggregate. There is almost 4 times increase in the strength of composite ground compared with virgin clay. Making the zero-value material more useful by creating a circular economy
Exploring Passenger Loyalty: A Perspective Through Expectancy Theory in Public Transport Usage
Full text linkThe paper discusses the application of Expectancy Theory in understanding and improving passenger loyalty in public transportation systems. Expectancy Theory, originally developed by Victor Vroom in 1964, is a psychological framework that explains how individuals\u27 expectations influence their decision-making processes and behavior. The paper emphasizes the importance of three key components of the theory: expectancy (the belief that effort will lead to performance), instrumentality (the belief that performance will result in desired outcomes) and valence (the value placed on those outcomes). By applying Expectancy Theory to public transportation, it explores how aligning service quality with passenger expectations can enhance passenger satisfaction and loyalty. Factors such as comfort, responsiveness, facilities, safety, ticketing systems and information provision are identified as critical components of service quality that influence passenger experiences. The paper argues that public transportation providers can improve passenger loyalty by ensuring that passengers perceive a clear connection between their efforts (e.g., choosing a service) and the rewards (e.g., reliable and comfortable transportation). It concludes that a strategic approach that incorporates Expectancy Theory can lead to systematic improvements in public transportation, resulting in higher passenger satisfaction, increased loyalty, and more efficient use of resources. This approach not only benefits transportation providers but also enhances the overall passenger experience and contributes to the development of more reliable and user-friendly public transportation systems
Technical and Vocational Education and Training College Lecturers and 21st-Century Skills: Awareness and Implications for Teaching Practices
Full text linkThe increasing demand for 21st-century skills in the workforce places new expectations on educators, including TVET lecturers, who play a critical role in preparing students for evolving industry needs. This study explored the awareness and understanding of 21st-century skills among TVET lecturers and examined how this awareness influences their teaching practices. A mixed-methods approach was employed, incorporating surveys, interviews, classroom observations, and document analysis. Results indicate that while lecturers are aware of critical skills such as communication, collaboration, creativity, and critical thinking, their practical application in teaching practices varies due to challenges such as curriculum rigidity, lack of resources, and insufficient professional development opportunities. The study concludes with strategic recommendations for professional development and curriculum adjustments necessary to foster 21st-century skills among lecturers.
Exploring The Relationship Between Personality Types and Learning Styles among Special Education Students in TVET: A Case Study
No full textThis study investigated how student personality types (introvert, extrovert, and ambivert) and learning styles, based on the five stimuli in the Dunn & Dunn Model, impacted special education students\u27 interest in Technical and Vocational Education and Training (TVET) skills related to refrigeration and air conditioning. A quantitative research method was employed, focusing on students enrolled in refrigeration and air conditioning programmes at Vocational Special Education Secondary Schools in Malaysia. A total of 37 special education students participated in this study, which utilized a questionnaire as a research tool. The study examined how introverted, extroverted, and ambivert personalities influenced students\u27 interest in the field. Additionally, it explored how various components of learning styles, including environment, emotions, sociology, physical aspects, and psychology, affected student interest. The findings indicated that most respondents had a moderately high level of individual personality classifications (introvert, extrovert, and ambivert) and a high-level learning style. The results revealed no significant link between different personality types and students\u27 interest in TVET skills related to refrigeration and air conditioning, with a correlation coefficient of 0.296. Various personality types did not notably affect special education students\u27 interest in the profession. This study enhanced comprehension of the interests of special education students, which could contribute to developing a more suitable and comprehensive learning approach to support student achievement in TVET skill
The Kansei Engineering Method for Determining the Emotional Design Concept of the Metaverse Library Interface: Technical University Students Perspectives
No full textThe rise of metaverse technology has impacted the way education is experienced through immersive interfaces, making user interface (UI) design a critical factor in shaping its emotional impact on users; the key to fostering engagement and satisfaction. However, providing emotion-centric UI that effectively satisfy user demands and technological competencies is an ongoing challenge. For that reason, this study examines the emotional responses from students of a technical university regarding metaverse library interfaces, aiming to identify the mapping correlation between UI design factors and the user-demanded experience. This experiment employed Kansei Engineering (KE) method, incorporating 28 specimens and 60 Kansei Words (KWs). Through the application of multivariate statistical analysis, the emotionally demanding UI concepts for the metaverse library were determined, with “radiance”, “orderliness”, and “conciseness” identified as the most significant (72.68%). Drawing from these findings, this study provides valuable guidance for improving user experience (UX) and shaping future design strategies of the metaverse libraries in technical universities. This research not only enhances emotional experience through UI design but also contributes to the growing importance of digital literacy in modern technical-based education and educational technology development, aligned with the goal of Technical and Vocational Education and Training (TVET)