Browsing by Author "Rankothge, W"

Now showing 1 - 12 of 12

Open Access
Experimental study on an efficient dengue disease management system: Planning and optimizing hospital staff allocation
(The Science and Information (SAI) Organization, 2018-01) Jayasuriya, M. M. C; Galappaththi, G. K. K. T; Sampath, M. A. D; Nipunika, H. N; Rankothge, W
Dengue has become a serious health hazard in Sri Lanka with the increasing cases and loss of human lives. It is necessary to develop an efficient dengue disease management system which could predict the dengue outbreaks, plan the countermeasures accordingly and allocate resources for the countermeasures. We have proposed a platform for Dengue disease management with following modules: (1) a prediction module to predict the dengue outbreak and (2) an optimization algorithm module to optimize hospital staff according to the predictions made on future dengue patient counts. This paper focuses on the optimization algorithm module. It has been developed based on two approaches: (1) Genetic Algorithm (GA) and (2) Iterated Local Search (ILS). We are presenting the performances of our optimization algorithm module with a comparison of the two approaches. Our results show that the GA approach is much more efficient and faster than the ILS approach.
Embargo
Investigating the Performance in SDN Based Data Centers Under Different Network Topologies
(IEEE, 2022-02-23) Rankothge, W; Hemachandra, K. G. R. P.; Jayasena, K. P. N; Wijesiri, M. P. M
A data center architecture can be determined as the basic structure of a cloud computing data center. However, the expected outcome cannot be obtained even though the tools, technologies, and elements are advanced. It is crucial to select the right architecture as architecture used, which directly affects the operational cost and power consumption of the data center. The SDN, Software Defined Networking, is a novel technology of cloud computing. This SDN is used to separate the control panel and forwarding plane as well, as it enables programmable network components. Therefore, the components of the SDN can easily control and managed. This research aimed to select an SDN-based architecture for a data center. Here, there are two parameters of average throughput and average packet delay on the three existing architectures that are tested using simulation. Then the results are compared with each other. In this research, the accuracy correlations among average packet delay and average throughput are investigated. The Obtained results show that Fat-Tree architecture is the best one among VL2 and hybrid architecture consists with flat tree and VL2 technology in terms of their performances.
Open Access
Machine Learning Modelling of the Relationship between Weather and Paddy Yield in Sri Lanka
(Hindawi, 2021-05) Ekanayake, P; Rankothge, W; Weliwatta, R; Jayasinghe, J. M. J. W
This paper presents the development of crop-weather models for the paddy yield in Sri Lanka based on nine weather indices, namely, rainfall, relative humidity (minimum and maximum), temperature (minimum and maximum), wind speed (morning and evening), evaporation, and sunshine hours. The statistics of seven geographical regions, which contribute to about two-thirds of the country’s total paddy production, were used for this study. The significance of the weather indices on the paddy yield was explored by employing Random Forest (RF) and the variable importance of each of them was determined. Pearson’s correlation and Spearman’s correlation were used to identify the behavior of correlation in a positive or negative direction. Further, the pairwise correlation among the weather indices was examined. The results indicate that the minimum relative humidity and the maximum temperature during the paddy cultivation period are the most influential weather indices. Moreover, RF was used to develop a paddy yield prediction model and four more techniques, namely, Power Regression (PR), Multiple Linear Regression (MLR) with stepwise selection, forward (step-up) selection, and backward (step-down) elimination, were used to benchmark the performance of the machine learning technique. Their performances were compared in terms of the Root Mean Squared Error (RMSE), Correlation Coefficient (R), Mean Absolute Error (MAE), and the Mean Absolute Percentage Error (MAPE). As per the results, RF is a reliable and accurate model for the prediction of paddy yield in Sri Lanka, demonstrating a very high R of 0.99 and the least MAPE of 1.4%.
Embargo
Multi-Cloud Container Communication Using Software Defined Networks
(IEEE, 2019-12-05) Najath, M; M.S Fayas Akram, M. F. S; Ahamed, I. A; Rankothge, W
Over the last decade, cloud computing has changed the IT world, mainly in the ways of storing, accessing and securing the users' data. However, there is no single cloud management platform in the market that can be used to manage different clouds (ex: AWS and AZURE), together with container communication between the clouds, to cater different cloud user requirements. We have proposed a platform for cloud service providers that automates the multi-cloud management related services with following modules: (1) communication, (2) resource allocation and (3) log view. Our results show that, using our proposed modules, the multi-cloud management related services can be automated efficiently and reliably.
Embargo
On the scaling of virtualized network functions
(IEEE, 2019-05-20) Rankothge, W; Ramalhinho, H; Lobo, J
Offering Virtualized Network Functions (VNFs) as a service requires automation of cloud resource management to allocate cloud resources for the VNFs dynamically. Most of the existing solutions focus only on the initial resource allocation. However, the allocation of resources must adapt to dynamic traffic demands and support fast scaling mechanisms. There are three basic scaling models: vertical where re-scaling is achieved by changing the resources assigned to the VNF in the host server, horizontal where VNFs are replicated or removed to do rescaling, and migration where VNFs are moved to servers with more resources. In this paper, we present an Iterated Local Search (ILS) based framework for automation of resource reallocation that supports the three scaling models. We, then, use the framework to run experiments and compare the different scaling approaches, specifically how the optimization is affected by the scaling approach and the optimization objectives.
Embargo
Optimization of Customer-Friendly Manual Load Shedding System
(IEEE, 2019-12-05) Fernando, W. D. I; Rankothge, W; Perera, A. D. S; Dissanayake, S. J; De Silva, W. D. S
To maintain the supply and demand of electricity power, load-Shedding is one of the methods practiced by the energy suppliers to hold the power system balanced, when an energy deficit problem arises. Lacking a proper load shedding scheme will lead to system instability and it will cause serious system frequency decay. We have proposed a solution to optimize the manual load shedding schedule with the application of optimization techniques, specifically the Genetic Algorithms. We have considered current hold by all feeders throughout the country, and the time period of load shedding as main factors in the optimization model. Our results show that, using our proposed model, we can minimize the imbalance between the supply and demand of electricity by selecting the best feeder to be selected for load shedding under given constraints.
Embargo
Smart Platform for Film Shooting Management
(IEEE, 2019-12-06) Senarath, S. M. M. M; Perera, M. T. K; Viduranga, D. G. R; Wijayananda, H. M. C. S; Rankothge, W
Producing a movie involves difficult and time-consuming phases, specially, pre-production and production. It's a challenging task to find out suitable locations for each scene and building a schedule without any clashes. We have proposed and implemented a platform for film shooting management with following modules: (1) identify required background for each scene, (2) classify available film shooting locations, (3) compare the required background and available film shooting locations and (4) schedule the shooting of each scene. We have used natural language processing, image processing, string matching algorithms and optimization techniques to implement the above-mentioned modules. Our results show that, using our proposed modules, the film shooting management related services can be automated efficiently and effectively.
Embargo
Supply and Demand Planning of Electricity Power: A Comprehensive Solution
(IEEE, 2019-12-06) Perera, S; Dissanayake, S; Fernando, D; De Silva, S; Rankothge, W
Electrical energy is one of the fastest growing energy demands in the world. Uncertainty in supplying the demand can threaten the social economic aspects of a country. The biggest driver of electrical demand is weather. Climatic changes not only affect the demand but also renewable energy supply. Wind and Solar are two alternative energy sources with less pollution. We have proposed a platform which helps energy providers, energy traders with services related to electricity supply and demand planning, with following modules. (1) Forecasting electricity consumption patterns (2) Forecasting wind power generation (3) Optimizing Load Shedding. Our platform has been implemented using statistical and machine learning techniques: Multi-Linear Regression for consumption prediction, Random forest regression for wind power forecast, and genetic algorithm to optimize load shedding. Our results show that, using our proposed module, we can minimize the imbalance between the supply and demand of electricity by predicting the consumption patterns of consumers, predicting the wind power generation and by selecting the best feeder to be selected for load shedding under given constraints.
Open Access
Sustainable manufacturing: application of optimization to textile manufacturing plants
(Global Journals, 2020-10-21) Liyanage, I; Nuwanga, S; Anjana, R; Rankothge, W; Gamage, N
The main goal of manufacturing industry is to produce the end products on time with good quality and keep the resource wastage low. However, manufacturing industry face several challenges such as bottle necks in the workflow, unsynchronized production, and sudden increase in product demands.In this paper, we are proposing a management platform for textile manufacturing plants with following modules: (1) sewing workflow optimization (2) quality assurance workflow optimization and (3) finishing workflow optimizations. We have used Genetic Programming (GP) approach, to optimize the workflows, considering different factors that affect each workflow. Our results show that, using our proposed platform, the manufacturing workflows can be optimized and reduce the bottle necks in the workflows and resource wastage in the manufacturing plant.
Embargo
Sustainable tourism: Application of optimization algorithms to schedule tour plans
(IEEE, 2019-01-31) Perera, D; Rathnayaka, C; Siriweera, L; Dilan, S; Rankothge, W
One of the challenging problems in the tourism industry is to maintain the environmental sustainability of the tourists attracted locations while giving a better user experience for the tourists. The proposed platform for sustainable tourism management system consist with following modules: A prediction module to predict an approximate value on tourist arrival for each location, an optimization algorithm module to decide the number of tourists that can be accommodated in each location considering the environmental sustainability, and an optimal path generating module to show the best route to each location. The optimization algorithm module is developed to decide the number of tourists for each location based on two approaches: Genetic Algorithms and Iterated Local Search. Next the optimal path generating module is developed based on traveling salesman problem.In this paper, the performances of the optimization algorithm module and the optimal path generating module is presented. Results show that, using the suggestions given by the algorithms help the tourist to enjoy a better experience in travelling while ensuring the sustainability in the tourism industry.
Embargo
Towards a Smart City: Application of Optimization for a Smart Transportation Management System
(IEEE, 2018-12) Thiranjaya, C; Rushan, R; Udayanga, P; Kaushalya, U; Rankothge, W
Intelligent traffic planning, the efficiency of public transport and the improved connectivity of all road users in a city, comprise the mobility characteristics of a smart city. In the era of smart cities, efficient and well managed public transportation systems play a crucial role. The planning and allocation of public transportation systems, especially the public bus scheduling is one of the major resource allocation problems where the optimal resource allocation increases the passenger's as well as bus owner's satisfaction. In this research, we have proposed a platform for public transportation management, especially for optimal planning and scheduling of buses. We have used two approaches for our algorithms: Iterated Local Search (ILS) and Genetic Algorithm (GA). In this paper, we are presenting our optimization algorithms and their performances. Our results show that, using our algorithms, we can decide the optimal allocations of buses and plan the bus schedules dynamically in the order of seconds.
Embargo
Towards Smart Farming: Accurate Prediction of Paddy Harvest and Rice Demand
(IEEE, 2019-01-31) Hashini Saranga, A. M; Weerakkody, W. A. N. D; Palliyaguru, S. T; Muthusinghe, R; Rankothge, W
Rice is the predominant staple food in Asian countries. It has a major impact on the social and economic development of these countries. Therefore, it is very important to keep the sustainability between paddy cultivation and consumer demand. Paddy crop yield and demand for rice of a country depend on numerous factors such as rainfall, humidity, citizen's life styles etc. Hence, the prediction of future harvest and demand is a complex process. There is a requirement for a platform that predicts on future harvest and demands based on all affecting factors. We have proposed a platform that targets the smart farming concepts for paddy, with following modules: (1) a prediction module to predict paddy harvest and (2) a prediction module to predict rice demand. We have developed the prediction modules using two machine learning algorithms: (1) Recurrent Neural Network (RNN) and (2) Long Short-Term Memory (LSTM). The performances of algorithms were evaluated using real data sets for the Sri Lankan context. Our results show that the prediction modules are giving accurate results in a short time.