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: Thesis-Abstract
Study on land fragmentation in southern province, Sri Lanka ; based on landscape ecology theory
(2023) Jayathunga, JND; Jayasinghe, A
As the landscape is where the majority of economic and human activities take place, it is an effective geographical scale for investigating the long-term consequences of anthropogenic activities on the environment. Land use changes in Sothem province, Sri Lanka, have increased significantly during the last three decades due to increasing of human needs. This research aims to better understand the interplay between land use/cover change and landscape fragmentation in Sothern Province by analyzing the dynamics of these processes. The study comprised three key stages as quantify and examine land use/cover change on landscape structure in terms of fragmentation; examine spatial transformation processes in landscape; detect land use land cover (LULC) changes in the landscape from 1988 to 2021. For this purpose, the United States Geological Survey (USGS) uses satellite images taken in 1988, 2005, 2016, and 2021 to categorize land use and land cover. Arc GIS 10.3 and QGIS 3.28 were used to process satellite images and maximum likelihood classification method was employed. FRAGSTAT 4.2 and MS Excel used to analyze class level metrics of fragmentation and spatial transformation. Agricultural land, home gardens, paddy fields, and vegetative cover have all risen in size and number since 1988, leading to a steady rise in patch density (PD). The low large patch index (LSI) and rapid reduction in PD both point to a severely fragmented landscape. Agricultural land, paddy land, vegetation, and dense vegetation cover are all negatively affected by the fragmentation. From 2005 to 2021, there was a greater amount of disaggregation in the LSI of agricultural land, home gardens, and vegetation cover. That was clear evidence of the landscape’s increasing complexity. Further, perforation is the first step in the chain reaction that leads to vegetation fragmentation. Findings further show the agricultural areas have been indicated to have a positive link with the vegetation (correlation value 0.77139), while home gardens have been found to have a negative correlation (correlation value -0.81306). The results of this study provide a foundation for quantifying and analyzing land use change and land fragmentation, which in turn allows environmental planners to assess the current state of affairs, forecast possible future developments, and develop effective spatial planning strategies and land monitoring mechanisms in pursuit of sustainable development.
item: Thesis-Abstract
Real-time human detection analytics in constrained image inputs
(2022) Fernando, H; Perera , I; De Silva, C
by an increasing number of surveillance camera systems. Analyzing human movement can be used for the wellbeing of humans. There are a set of analytical tools and algorithms which can be used to detect, track, and analyze humans in images. Human movement analytics has various subdomains including human detection, human recognition, human tracking, human localization, human reidentification, human behavior analysis, and abnormal activity detection. Human detection is the most crucial step among them, and which helps to derive other sub domains. Human detection analytics in constrained lighting conditions would be a challenging task to apply due to the low contrast of the image context. Currently available systems focused on the daytime. The background light is an essential factor in the camera images, which rigorously affects the quality of the image. We can identify considerable differences if we compare two images at the rich light condition and constrained light condition. Fewer features of the objects can be extracted in constrained light conditions than rich light conditions. Illumination of the background context is an important factor if we focus on such applications. Currently, most researchers have used human detection analytics in visible light. RGB image shows a clear view when there is sufficient light existing, and it is highly sensitive to visible light conditions compared to infrared. In this research, we considered infrared images as constrained image inputs. Cte p posed methodology contains a novel human detection approach based on machine le&roiop and a motion dynamic model. Here we have addressed the problem using a combination of L -:0p Convolutional Neural Networks (DCNN) for human detection and Kemelized Correlation Filters (KCF) for human tracking. MobileNet pre-trained model is used for framewise human detection as the first step. Then the KCF object tracking algorithm is used to increase the human detection accuracy while tracking the human in the context. Furthermore, we applied some preprocessing techniques to reduce the noise effects. Currently, the progress made by this research-based project is sufficient to initiate the development of a complete human detection analysis solution based on live CCTV camera footage. This solution provides the core functionality of human detection analytics and it can be easily adapted to different domain solutions such as customer behavior analytics in a supermarket or worker movement analytics in an industrial premise.
item: Thesis-Abstract
Integrating machine learning to optimize stock flow and goods replenishment in intralogistics
(2022) Weerasinghe, WAKV; Perera, HN
The intralogistics area has emerged and developed as a result of increasing the importance of Industry 4.0. Autonomously running intralogistics systems are gaining more and more attention as a result of digitalization. Intralogistics is a process of organization, control, execution, and optimization of internal material and information movement. It is a complex interplay of several logistical operations. Hie literature study highlighted how logistics inside the warehouse facility transformed with the major trends in the context of optimization, automation, system integration, and mathematical modeling techniques. Further, this study proves that numerous technologies can be altered and combined to improve intralogistics in terms of accuracy, quality, efficacy as well as sustainable aspects even in more complex settings in the discussion. In the paradigm shift of moving traditional internal logistics systems to autonomous advanced intralogistics systems, integrating various operations research applications with industry 4.0 concepts is vital. It will support to develop optimized, organized, and automated logistic systems within the warehouse. In this study, we have considered developing a storage and goods retrieval system amalgamating the intralogistics concept. In order to develop a demand-driven optimized storage plan for the warehouse, storage allocation was focused. The storage allocation was done by classification technique mainly considering the total consumption value of the items. The used classification mechanism was improved by integrating Machine learning (ML) approach. ML-based classification addresses the question that how a computerized system can automatically identify the correct storage segmentation for the items in the inventory including newly added items. Apart from the consumption value, cost of the item issued quantity of the item, and frequency of issuing were considered to develop the model. It is easy to retrieve the items from the warehouse as it is in the correct demarcated location to fulfill the orders received for the warehouse. The developed storage allocation will be used to perform a simulation model in order to identify how inventory segmentation utilizes and eases the internal logistics.
item: Thesis-Abstract
Analysis of moisture removal efficiency and particle size development inside fluidized bed drying process at a full cream milk powder manufacturing plant in Sri Lanka
(2023) Jayawardane, HT; Rathnayake, HHMP
Milk powder manufacturing can be identified as one of the leading industrial processes in the global dairy sector for cow milk preservation with nutrition purposes and processing as a raw material for many other food processing applications. The global milk powder market has already reached 11.2 million tonnes in 2020 and is expected to reach a volume of 14.1 million in 2026. Milk Powder is manufactured by dehydrating liquid milk by removing free moisture with some amount of bound moisture and leaving only milk solids by falling film evaporation and multi-stage drying processes. Milk powder has an extremely long shelf life when compared to raw milk and does not need refrigeration. This manufacturing process includes many chemical engineering unit operations, such as falling film evaporation, flash evaporation, direct steam injection, thermal vapor recompression, spray drying, fluidized bed drying, dehumidification, cyclone separation, screening, mixing, centrifugal separation, pasteurization, homogenization, etc. Drying processes play a significant role in the desired removal of moisture which will help particle size improvements as well as final milk powder quality parameter improvements. Multi-stage drying processes are included in milk powder manufacturing for moisture removal. Three-stage dryers, as well as two-stage dryers are widely used in the process industry to achieve desired moisture removal with improved quality parameters. The two-stage drying process, which is a combination of spray drying and fluidized bed drying processes are widely used in Sri Lankan milk powder manufacturing plant operations. Among these two drying processes, the fluidized bed drying process will directly help to improve the particle diameter as well as desired final milk powder quality parameters like wettability, insolubility index, bulk density, etc. In this study, the existing literature on the milk powder manufacturing process and fluidized bed dryer operations, including moisture removal were reviewed and analysed. In this study, the moisture removal process of a fluidized bed dryer is analysed with the aim of identifying the reason behind less solubility of produced full cream milk powder in hot water. Fluidized bed dryer operation can be identified as the main unit operation of the entire manufacturing process which directly helps to improve the final product quality parameters like particle diameter, insolubility index, wettability, bulk density, etc. To analyse the moisture removal process of full cream milk powder, samples were collected from the inlet of the fluidized bed dryer, from the heating section, and from the cooling section, as well as the full cream milk powder outlet. Moisture removal activity is compared with the design parameters of the equipment and reviewed the moisture removal gradient to predict the improvement of particle diameter which directly affect the solubility characteristics. In addition, final milk powder product particle size improvement in each section of the fluidized bed dryer were analyzed and compared with the literature values to have a sound idea about quality improvements. This study shows that the moisture removal process is the key process operation, which determines the final product quality related to the solubility and particle size development of full cream milk powder. The study identifies a relationship between the moisture removal process as well as gradual particle size development along the fluidized bed that would support the related industries in Sri Lanka to improve the full-cream milk powder manufacturing process.
item: Thesis-Abstract
Exploring the contribution of day and night street markets to urban liveliness and vibrancy : a case study of the Pamunuwa day/night market in Maharagama town
(2024) Bandara, MGGN; Dharmasena, J
Urban areas derive their identity not solely from fixed structures but also from dynamic elements that imbue cities with uniqueness and character. To maintain a constant influx of people throughout the day and night, cities require sustained activities beyond seasonal events. Street markets function as perpetual attractions, drawing individuals into urban spaces at all hours and nurturing vibrancy. Despite their potential to significantly impact urban liveliness, limited research has delved into the specific contributions of street markets to the urban fabric. This study examines the positive contributions of the Pamunuwa day and night street market to the liveliness and vibrancy of Maharagama town. While the daytime street market is well-known, the night market remains relatively unexplored despite evolving into a thriving economic and social hub, attracting visitors from across the country. The public venue increases informal surveillance, transforming the urban context into a safe, vibrant space with diverse uses. The study outlines the essential components of urban vibrancy and liveliness as related to day and night street markets, their contribution, and impacts, and establishes effective parameters for their measurement. The study’s findings prove that the Pamunuwa day and night street market significantly impacts the Maharagama town’s overall liveliness and vibrancy.