Search results for: multiple arc network model

Authors: Albert W. K. Tan

Abstract:

This research consolidates a comprehensive array of publications from peer-reviewed journals, case studies, and seminar reports focused on reverse logistics and network design. By synthesizing this secondary knowledge, our objective is to identify and articulate key decision factors crucial to reverse logistics network design for e-commerce. Through this exploration, we aim to present a refined mathematical model that offers valuable insights for companies seeking to optimize their reverse logistics operations. The primary goal of this research endeavor is to develop a comprehensive framework tailored to advising organizations and companies on crafting effective networks for their reverse logistics operations, thereby facilitating the achievement of their organizational goals. This involves a thorough examination of various network configurations, weighing their advantages and disadvantages to ensure alignment with specific business objectives. The key objectives of this research include: (i) Identifying pivotal factors pertinent to network design decisions within the realm of reverse logistics across diverse supply chains. (ii) Formulating a structured framework designed to offer informed recommendations for sound network design decisions applicable to relevant industries and scenarios. (iii) Propose a mathematical model to optimize its reverse logistics network. A conceptual framework for designing a reverse logistics network has been developed through a combination of insights from the literature review and information gathered from company websites. This framework encompasses four key stages in the selection of reverse logistics operations modes: (1) Collection, (2) Sorting and testing, (3) Processing, and (4) Storage. Key factors to consider in reverse logistics network design: I) Centralized vs. decentralized processing: Centralized processing, a long-standing practice in reverse logistics, has recently gained greater attention from manufacturing companies. In this system, all products within the reverse logistics pipeline are brought to a central facility for sorting, processing, and subsequent shipment to their next destinations. Centralization offers the advantage of efficiently managing the reverse logistics flow, potentially leading to increased revenues from returned items. Moreover, it aids in determining the most appropriate reverse channel for handling returns. On the contrary, a decentralized system is more suitable when products are returned directly from consumers to retailers. In this scenario, individual sales outlets serve as gatekeepers for processing returns. Considerations encompass the product lifecycle, product value and cost, return volume, and the geographic distribution of returns. II) In-house vs. third-party logistics providers: The decision between insourcing and outsourcing in reverse logistics network design is pivotal. In insourcing, a company handles the entire reverse logistics process, including material reuse. In contrast, outsourcing involves third-party providers taking on various aspects of reverse logistics. Companies may choose outsourcing due to resource constraints or lack of expertise, with the extent of outsourcing varying based on factors such as personnel skills and cost considerations. Based on the conceptual framework, the authors have constructed a mathematical model that optimizes reverse logistics network design decisions. The model will consider key factors identified in the framework, such as transportation costs, facility capacities, and lead times. The authors have employed mixed LP to find the optimal solutions that minimize costs while meeting organizational objectives.

Keywords: reverse logistics, supply chain management, optimization, e-commerce

Procedia PDF Downloads 35

Authors: Zouhour Neji Ben Salem

Abstract:

Microscopic evacuation focuses on the evacuee behavior and way of search of safety place in an egress situation. In recent years, several models handled microscopic evacuation problem. Among them, we have proposed Artificial Neural Network (ANN) as an alternative to mathematical models that can deal with such problem. In this paper, we present two ANN models: SOM map and Hopfield Network used to predict the evacuee behavior in a disaster situation. These models are tested in a real case, the second floor of Tunisian children hospital evacuation in case of fire. The two models are studied and compared in order to evaluate their performance.

Keywords: artificial neural networks, self-organization map, hopfield network, microscopic evacuation, fire building evacuation

Procedia PDF Downloads 401

Authors: Danuvasin Charoen

Abstract:

The study investigates the business process and business model of GRABTAXI. The paper also discusses how the company implemented strategies to gain competitive advantages. The data is derived from the analysis of secondary data and the in-depth interviews among staffs, taxi drivers, and key customers. The findings indicated that the company’s competitive advantages come from being the first mover, emphasising on the ease of use and tangible benefits of application, and using network effect strategy.

Keywords: taxi, mobile application, innovative business model, Thailand

Procedia PDF Downloads 296

Authors: David Hasurungan

Abstract:

This paper objective is to analyze the impact from photovoltaic system integration to power distribution network. The case study in Badak Liquefied Natural Gas (LNG) plant is presented in this paper. Badak LNG electricity network is operated in islanded mode. The total power generation in Badak LNG plant is significantly affected to feed gas supply. Meanwhile, to support the Government regulation, Badak LNG continuously implemented the grid-connected photovoltaic system in existing power distribution network. The impact between train operational mode change in Badak LNG plant and the growth of photovoltaic system is also encompassed in analysis. The analysis and calculation are performed using software Power Factory 15.1.

Keywords: power quality, distribution network, grid-connected photovoltaic system, power management system

Procedia PDF Downloads 356

Authors: Thomas E. Portegys

Abstract:

An animal behavior problem is presented in the form of a nest-building task that involves two cooperating virtual birds, a male and female. The female builds a nest into which she lays an egg. The male's job is to forage in a forest for food for both himself and the female. In addition, the male must fetch stones from a nearby desert for the female to use as nesting material. The task is completed when the nest is built, and an egg is laid in it. A goal-seeking neural network and a recurrent neural network were trained and tested with little success. The goal-seeking network was then enhanced with “place cells”, allowing the birds to spatially navigate the world, building the nest while keeping themselves fed. Place cells are neurons in the hippocampus that map space.

Keywords: artificial animal intelligence, artificial life, goal-seeking neural network, nest-building, place cells, spatial navigation

Procedia PDF Downloads 53

Authors: Ying Su, Morgan C. Wang

Abstract:

Long-term time series forecasting is an important research area for automated machine learning (AutoML). Currently, forecasting based on either machine learning or statistical learning is usually built by experts, and it requires significant manual effort, from model construction, feature engineering, and hyper-parameter tuning to the construction of the time series model. Automation is not possible since there are too many human interventions. To overcome these limitations, this article proposed to use recurrent neural networks (RNN) through the memory state of RNN to perform long-term time series prediction. We have shown that this proposed approach is better than the traditional Autoregressive Integrated Moving Average (ARIMA). In addition, we also found it is better than other network systems, including Fully Connected Neural Networks (FNN), Convolutional Neural Networks (CNN), and Nonpooling Convolutional Neural Networks (NPCNN).

Keywords: automated machines learning, autoregressive integrated moving average, neural networks, time series analysis

Procedia PDF Downloads 100

Authors: Jayanta Kumar Ghosh, Swastik Sunil Goriwale, Himangshu Sarkar

Abstract:

Groundwater is one of the most valuable natural resources that society consumes for its domestic, industrial, and agricultural water supply. Its bulk and indiscriminate consumption affects the groundwater resource. Often, it has been found that the groundwater recharge rate is much lower than its demand. Thus, to maintain water and food security, it is necessary to monitor and management of groundwater storage. However, it is challenging to estimate groundwater storage (GWS) by making use of existing hydrological models. To overcome the difficulties, machine learning (ML) models are being introduced for the evaluation of groundwater level (GWL). Thus, the objective of this research work is to develop an ML-based model for the prediction of GWL. This objective has been realized through the development of an artificial neural network (ANN) model based on hydro-gravimetry. The model has been developed using training samples from field observations spread over 8 months. The developed model has been tested for the prediction of GWL in an observation well. The root means square error (RMSE) for the test samples has been found to be 0.390 meters. Thus, it can be concluded that the hydro-gravimetric-based ANN model can be used for the prediction of GWL. However, to improve the accuracy, more hydro-gravimetric parameter/s may be considered and tested in future.

Keywords: machine learning, hydro-gravimetry, ground water level, predictive model

Procedia PDF Downloads 121

Authors: M. J. Park, S. H. Lee, C. H. Lee, O. M. Kwon

Abstract:

This paper studies a secondary voltage control framework of the microgrids based on the consensus for a communication network of multiagent. The proposed control is designed by the communication network with one-way links. The communication network is modeled by a directed graph. At this time, the concept of sampling is considered as the communication constraint among each distributed generator in the microgrids. To analyze the sampling effects on the secondary voltage control of the microgrids, by using Lyapunov theory and some mathematical techniques, the sufficient condition for such problem will be established regarding linear matrix inequality (LMI). Finally, some simulation results are given to illustrate the necessity of the consideration of the sampling effects on the secondary voltage control of the microgrids.

Keywords: microgrids, secondary control, multiagent, sampling, LMI

Procedia PDF Downloads 329

Authors: Vahid Nourani, Atefeh Ashrafi

Abstract:

Prediction of treated wastewater quality is a matter of growing importance in water treatment procedure. In this way artificial neural network (ANN), as a robust data-driven approach, has been widely used for forecasting the effluent quality of wastewater treatment. However, developing ANN model based on appropriate input variables is a major concern due to the numerous parameters which are collected from treatment process and the number of them are increasing in the light of electronic sensors development. Various studies have been conducted, using different clustering methods, in order to classify most related and effective input variables. This issue has been overlooked in the selecting dominant input variables among wastewater treatment parameters which could effectively lead to more accurate prediction of water quality. In the presented study two ANN models were developed with the aim of forecasting effluent quality of Tabriz city’s wastewater treatment plant. Biochemical oxygen demand (BOD) was utilized to determine water quality as a target parameter. Model A used Principal Component Analysis (PCA) for input selection as a linear variance-based clustering method. Model B used those variables identified by the mutual information (MI) measure. Therefore, the optimal ANN structure when the result of model B compared with model A showed up to 15% percent increment in Determination Coefficient (DC). Thus, this study highlights the advantage of PCA method in selecting dominant input variables for ANN modeling of wastewater plant efficiency performance.

Keywords: Artificial Neural Networks, biochemical oxygen demand, principal component analysis, mutual information, Tabriz wastewater treatment plant, wastewater treatment plant

Procedia PDF Downloads 125

Authors: Paul C. Njoku, Archana Swati Njoku

Abstract:

The study deals with the development of a mathematical model to characterize the oil production in Nigeria. This is calculated by initiating the dynamics of oil production in million barrels revenue plan cost of oil production in million nairas and unit cost of production from 1974-1982 in the contest of the federal Republic of Nigeria. This country export oil to other countries as well as importing specialized crude. The transport network from origin/destination tij to pairs is taking into account simulation runs, optimization have been considered in this study.

Keywords: mathematical oil model development dynamics, Nigeria, characterization barrels, dynamics of oil production

Procedia PDF Downloads 384

Authors: Francesco Buccafurri, Celeste Romolo

Abstract:

Nowadays phishing is the most frequent starting point of cyber-attack vectors. Phishing is implemented both via email and social network messages. While a wide scientific literature exists which addresses the problem of contrasting email spam-phishing, no specific countermeasure has been so far proposed for phishing included into private messages of social network platforms. Unfortunately, the problem is severe. This paper proposes an approach against social network phishing, based on a non invasive collaborative information-sharing approach which leverages blockchain. The detection method works by filtering candidate messages, by distilling them by means of a distance-preserving hash function, and by publishing hashes over a public blockchain through a trusted smart contract (thus avoiding denial of service attacks). Phishing detection exploits social information embedded into social network profiles to identify similar messages belonging to disjoint contexts. The main contribution of the paper is to introduce a new approach to contrasting the problem of social network phishing, which, despite its severity, received little attention by both research and industry.

Keywords: phishing, social networks, information sharing, blockchain

Procedia PDF Downloads 325

Authors: M. H. Fazel Zarandi, N. Moshahedi

Abstract:

The hub location problem appears in a variety of applications such as medical centers, firefighting facilities, cargo delivery systems and telecommunication network design. The location of service centers has a strong influence on the congestion at each of them, and, consequently, on the quality of service. This paper presents a fuzzy maximal hub covering location problem (FMCHLP) in which travel costs between any pair of nodes is considered as a fuzzy variable. In order to consider the quality of service, we model each hub as a queue. Arrival rate follows Poisson distribution and service rate follows Erlang distribution. In this paper, at first, a nonlinear mathematical programming model is presented. Then, we convert it to the linear one. We solved the linear model using GAMS software up to 25 nodes and for large sizes due to the complexity of hub covering location problems, and simulated annealing algorithm is developed to solve and test the model. Also, we used possibilistic c-means clustering method in order to find an initial solution.

Keywords: fuzzy modeling, location, possibilistic clustering, queuing

Procedia PDF Downloads 390

Authors: Xuan Tran

Abstract:

The purpose of this paper is to present a comprehensive dynamic model for pricing strategies in the hotel competition to find a win-win situation for the competitive set. By utilizing the Cobb-Douglas utility model, the study establishes room rates by analyzing the price elasticity of demand across a competitive set of four hotels, with a focus on occupancy rates. To further enhance the analysis, game theory is applied to identify the best response for each competitive party, which illustrates the optimal pricing strategy for each hotel in the competitive landscape. This approach offers valuable insights into how hotels can strategically adjust their room rates in response to market conditions and competitor actions. The primary contributions of this research include as follows: (1) advantages for both individual hotels and the broader competitive hotel market, (2) benefits for hotel management overseeing multiple brands, and (3) positive impacts on the local community.

Keywords: dynamic model, game theory, best response, Cobb-Douglas

Procedia PDF Downloads 17

Authors: Yuanjin Liu

Abstract:

Retirement withdrawal strategies are very important to minimize the probability of ruin in retirement. The ruin probability is modeled as a function of initial withdrawal age, gender, asset allocation, inflation rate, and initial withdrawal rate. The ruin probability is obtained based on the 2019 period life table for the Social Security, IRS Required Minimum Distribution (RMD) Worksheets, US historical bond and equity returns, and inflation rates using simulation. Several popular machine learning algorithms of the generalized additive model, random forest, support vector machine, extreme gradient boosting, and artificial neural network are built. The model validation and selection are based on the test errors using hyperparameter tuning and train-test split. The optimal model is recommended for retirees to monitor the ruin probability. The optimal withdrawal strategy can be obtained based on the optimal predictive model.

Keywords: ruin probability, retirement withdrawal strategies, predictive models, optimal model

Procedia PDF Downloads 71

Authors: Piotr A. Kowalski, Kasper Sapala, Wiktor Warchalowski

Abstract:

PM10 is a suspended dust that primarily has a negative effect on the respiratory system. PM10 is responsible for attacks of coughing and wheezing, asthma or acute, violent bronchitis. Indirectly, PM10 also negatively affects the rest of the body, including increasing the risk of heart attack and stroke. Unfortunately, Poland is a country that cannot boast of good air quality, in particular, due to large PM concentration levels. Therefore, based on the dense network of Airly sensors, it was decided to deal with the problem of prediction of suspended particulate matter concentration. Due to the very complicated nature of this issue, the Machine Learning approach was used. For this purpose, Convolution Neural Network (CNN) neural networks have been adopted, these currently being the leading information processing methods in the field of computational intelligence. The aim of this research is to show the influence of particular CNN network parameters on the quality of the obtained forecast. The forecast itself is made on the basis of parameters measured by Airly sensors and is carried out for the subsequent day, hour after hour. The evaluation of learning process for the investigated models was mostly based upon the mean square error criterion; however, during the model validation, a number of other methods of quantitative evaluation were taken into account. The presented model of pollution prediction has been verified by way of real weather and air pollution data taken from the Airly sensor network. The dense and distributed network of Airly measurement devices enables access to current and archival data on air pollution, temperature, suspended particulate matter PM1.0, PM2.5, and PM10, CAQI levels, as well as atmospheric pressure and air humidity. In this investigation, PM2.5, and PM10, temperature and wind information, as well as external forecasts of temperature and wind for next 24h served as inputted data. Due to the specificity of the CNN type network, this data is transformed into tensors and then processed. This network consists of an input layer, an output layer, and many hidden layers. In the hidden layers, convolutional and pooling operations are performed. The output of this system is a vector containing 24 elements that contain prediction of PM10 concentration for the upcoming 24 hour period. Over 1000 models based on CNN methodology were tested during the study. During the research, several were selected out that give the best results, and then a comparison was made with the other models based on linear regression. The numerical tests carried out fully confirmed the positive properties of the presented method. These were carried out using real ‘big’ data. Models based on the CNN technique allow prediction of PM10 dust concentration with a much smaller mean square error than currently used methods based on linear regression. What's more, the use of neural networks increased Pearson's correlation coefficient (R²) by about 5 percent compared to the linear model. During the simulation, the R² coefficient was 0.92, 0.76, 0.75, 0.73, and 0.73 for 1st, 6th, 12th, 18th, and 24th hour of prediction respectively.

Keywords: air pollution prediction (forecasting), machine learning, regression task, convolution neural networks

Procedia PDF Downloads 145

Authors: Muteb Alotaibi, Graham Clarke, Nick Malleson

Abstract:

The Hajj religious festival at Mina in Saudi Arabia has always presented the opportunity for injuries or deaths. For example, in 1990, a stampede killed 1426 pilgrims, whilst in 1997, 343 people were killed and 1500 injured due to a fire fuelled by high winds sweeping through the tent city in Mina.Many more minor incidents have occurred since then. It is predicted that 5 million pilgrims will soon perform the ritual at Mina (which is, in effect, a temporary city built each year in the desert), which might lead in the future to severe congestion and accidents unless the research is conducted on actions that contribute positively to improving the management of the crowd and facilitating the flow of pilgrims safely and securely. To help prevent further disasters, it is important to first plan better, more accessible locations for emergency services across Mina to ensure a good service for pilgrims. In this paper, we first use a Location Allocation Model (LAM) within a network GIS to examine the optimal locations for key services in the temporary city of Mina. This has been undertaken in relation to the location and movement of the pilgrims during the six day religious festival. The results of various what-if scenarios have been compared against the current location of services. A major argument is that planners should be flexible and locate facilities at different locations throughout the day and night. The use of location-allocation models in this type of comparative static mode has rarely been operationalised in the literature. Second, we model pilgrim movements and behaviours along with the most crowded parts of the network. This has been modelled using an agent-based model. This model allows planners to understand the key bottlenecks in the network and at what usage levels the paths become critically congested. Thus the paper has important implications and recommendations for future disaster planning strategies. This will enable planners to see at what stage in the movements of pilgrims problems occur in terms of potential crushes and trampling incidents. The main application of this research was only customised for pedestrians as the concentration only for pedestrians who move to Jamarat via foot. Further, the network in the middle of Mina was only dedicated for pedestrians for safety, so no Buses, trains and private cars were allowed in this area to prevent the congestion within this network. Initially, this research focus on Mina city as ‘temporary city’ and also about service provision in temporary cities, which is not highlighted in literature so far. Further, it is the first study which use the dynamic demand to optimise the services in the case of day and night time. Moreover, it is the first study which link the location allocation model for optimising services with ABM to find out whether or not the service location is located in the proper location in which it’s not affecting on crowd movement in mainstream flow where some pilgrims need to have health services.

Keywords: ABM, crowd management, hajj, temporary city

Procedia PDF Downloads 118

Authors: Susana Marques

Abstract:

Research on how store environment cues influence consumers’ store choice decision criteria, such as store operations, product quality, monetary price, store image and sales promotion, is sparse. Especially absent research on the simultaneous impact of multiple store environment cues. The authors propose a comprehensive store choice model that includes: three types of store environment cues as exogenous constructs; various store choice criteria as possible mediating constructs, and store patronage intentions as an endogenous construct. On the basis of testing with a sample of 561 customers of hypermarkets, the model is partially supported. This study used structural equation modelling to test the proposed model.

Keywords: store choice, store patronage, structural equation modelling, retailing

Procedia PDF Downloads 265

Authors: Suraj Mehrotra

Abstract:

The stock market is a volatile, bustling marketplace that is a cornerstone of economics. It defines whether companies are successful or in spiral. A thorough understanding of it is important - many companies have whole divisions dedicated to analysis of both their stock and of rivaling companies. Linking the world of finance and artificial intelligence (AI), especially the stock market, has been a relatively recent development. Predicting how stocks will do considering all external factors and previous data has always been a human task. With the help of AI, however, machine learning models can help us make more complete predictions in financial trends. Taking a look at the stock market specifically, predicting the open, closing, high, and low prices for the next day is very hard to do. Machine learning makes this task a lot easier. A model that builds upon itself that takes in external factors as weights can predict trends far into the future. When used effectively, new doors can be opened up in the business and finance world, and companies can make better and more complete decisions. This paper explores the various techniques used in the prediction of stock prices, from traditional statistical methods to deep learning and neural networks based approaches, among other methods. It provides a detailed analysis of the techniques and also explores the challenges in predictive analysis. For the accuracy of the testing set, taking a look at four different models - linear regression, neural network, decision tree, and naïve Bayes - on the different stocks, Apple, Google, Tesla, Amazon, United Healthcare, Exxon Mobil, J.P. Morgan & Chase, and Johnson & Johnson, the naïve Bayes model and linear regression models worked best. For the testing set, the naïve Bayes model had the highest accuracy along with the linear regression model, followed by the neural network model and then the decision tree model. The training set had similar results except for the fact that the decision tree model was perfect with complete accuracy in its predictions, which makes sense. This means that the decision tree model likely overfitted the training set when used for the testing set.

Keywords: machine learning, testing set, artificial intelligence, stock analysis

Procedia PDF Downloads 92

Authors: Venkat S. Somayajula

Abstract:

Ground surface temperature history prediction model plays a vital role in determining standards for international nuclear waste management. International standards for borehole based nuclear waste disposal require paleoclimate cycle predictions on scale of a million forward years for the place of waste disposal. This research focuses on developing a paleoclimate cycle prediction model using Bayesian long-short term memory (LSTM) neural architecture operated on accumulated borehole temperature history data. Bayesian models have been previously used for paleoclimate cycle prediction based on Monte-Carlo weight method, but due to limitations pertaining model coupling with certain other prediction networks, Bayesian models in past couldn’t accommodate prediction cycle’s over 1000 years. LSTM has provided frontier to couple developed models with other prediction networks with ease. Paleoclimate cycle developed using this process will be trained on existing borehole data and then will be coupled to surface temperature history prediction networks which give endpoints for backpropagation of LSTM network and optimize the cycle of prediction for larger prediction time scales. Trained LSTM will be tested on past data for validation and then propagated for forward prediction of temperatures at borehole locations. This research will be beneficial for study pertaining to nuclear waste management, anthropological cycle predictions and geophysical features

Keywords: Bayesian long-short term memory neural network, borehole temperature, ground surface temperature history, paleoclimate cycle

Procedia PDF Downloads 126

Authors: Deokho Kim, Wonwoo Lee, Jaewoong Lee, Teresa Ng, Gun-Ill Lee, Jiwon Jeong

Abstract:

This paper presents a method for improving object search accuracy using a deep learning model. A major limitation to provide accurate similarity with deep learning is the requirement of huge amount of data for training pairwise similarity scores (metrics), which is impractical to collect. Thus, similarity scores are usually trained with a relatively small dataset, which comes from a different domain, causing limited accuracy on measuring similarity. For this reason, this paper proposes a deep learning model that can be trained with a significantly small amount of data, a clustered data which of each cluster contains a set of visually similar images. In order to measure similarity distance with the proposed method, visual features of two images are extracted from intermediate layers of a convolutional neural network with various pooling methods, and the network is trained with pairwise similarity scores which is defined zero for images in identical cluster. The proposed method outperforms the state-of-the-art object similarity scoring techniques on evaluation for finding exact items. The proposed method achieves 86.5% of accuracy compared to the accuracy of the state-of-the-art technique, which is 59.9%. That is, an exact item can be found among four retrieved images with an accuracy of 86.5%, and the rest can possibly be similar products more than the accuracy. Therefore, the proposed method can greatly reduce the amount of training data with an order of magnitude as well as providing a reliable similarity metric.

Keywords: visual search, deep learning, convolutional neural network, machine learning

Procedia PDF Downloads 213

Authors: Jonathan Gong

Abstract:

Purpose AI-driven solutions are at the forefront of many pathology and medical imaging methods. Using algorithms designed to better the experience of medical professionals within their respective fields, the efficiency and accuracy of diagnosis can improve. In particular, X-rays are a fast and relatively inexpensive test that can diagnose diseases. In recent years, X-rays have not been widely used to detect and diagnose COVID-19. The under use of Xrays is mainly due to the low diagnostic accuracy and confounding with pneumonia, another respiratory disease. However, research in this field has expressed a possibility that artificial neural networks can successfully diagnose COVID-19 with high accuracy. Models and Data The dataset used is the COVID-19 Radiography Database. This dataset includes images and masks of chest X-rays under the labels of COVID-19, normal, and pneumonia. The classification model developed uses an autoencoder and a pre-trained convolutional neural network (DenseNet201) to provide transfer learning to the model. The model then uses a deep neural network to finalize the feature extraction and predict the diagnosis for the input image. This model was trained on 4035 images and validated on 807 separate images from the ones used for training. The images used to train the classification model include an important feature: the pictures are cropped beforehand to eliminate distractions when training the model. The image segmentation model uses an improved U-Net architecture. This model is used to extract the lung mask from the chest X-ray image. The model is trained on 8577 images and validated on a validation split of 20%. These models are calculated using the external dataset for validation. The models’ accuracy, precision, recall, f1-score, IOU, and loss are calculated. Results The classification model achieved an accuracy of 97.65% and a loss of 0.1234 when differentiating COVID19-infected, pneumonia-infected, and normal lung X-rays. The segmentation model achieved an accuracy of 97.31% and an IOU of 0.928. Conclusion The models proposed can detect COVID-19, pneumonia, and normal lungs with high accuracy and derive the lung mask from a chest X-ray with similarly high accuracy. The hope is for these models to elevate the experience of medical professionals and provide insight into the future of the methods used.

Keywords: artificial intelligence, convolutional neural networks, deep learning, image processing, machine learning

Procedia PDF Downloads 126

Authors: Jingsi Li, Neil S. Ferguson

Abstract:

Time pressure could have an influence on the productivity, quality of decision making, and the efficiency of problem-solving. This has been mostly stemmed from cognitive research or psychological literature. However, a salient scarce discussion has been held for transport adjacent fields. It is conceivable that in many activity-travel contexts, time pressure is a potentially important factor since an excessive amount of decision time may incur the risk of late arrival to the next activity. The activity-travel rescheduling behavior is commonly explained by costs and benefits of factors such as activity engagements, personal intentions, social requirements, etc. This paper hypothesizes that an additional factor of perceived time pressure could affect travelers’ rescheduling behavior, thus leading to an impact on travel demand management. Time pressure may arise from different ways and is assumed here to be essentially incurred due to travelers planning their schedules without an expectation of unforeseen elements, e.g., transport disruption. In addition to a linear-additive utility-maximization model, the less computationally compensatory heuristic models are considered as an alternative to simulate travelers’ responses. The paper will contribute to travel behavior modeling research by investigating the following questions: how to measure the time pressure properly in an activity-travel day plan context? How do travelers reschedule their plans to cope with the time pressure? How would the importance of the activity affect travelers’ rescheduling behavior? What will the behavioral model be identified to describe the process of making activity-travel rescheduling decisions? How do these identified coping strategies affect the transport network? In this paper, a Mixed Heuristic Model (MHM) is employed to identify the presence of different choice heuristics through a latent class approach. The data about travelers’ activity-travel rescheduling behavior is collected via a web-based interactive survey where a fictitious scenario is created comprising multiple uncertain events on the activity or travel. The experiments are conducted in order to gain a real picture of activity-travel reschedule, considering the factor of time pressure. The identified behavioral models are then integrated into a multi-agent transport simulation model to investigate the effect of the rescheduling strategy on the transport network. The results show that an increased proportion of travelers use simpler, non-compensatory choice strategies instead of compensatory methods to cope with time pressure. Specifically, satisfying - one of the heuristic decision-making strategies - is adopted commonly since travelers tend to abandon the less important activities and keep the important ones. Furthermore, the importance of the activity is found to increase the weight of negative information when making trip-related decisions, especially route choices. When incorporating the identified non-compensatory decision-making heuristic models into the agent-based transport model, the simulation results imply that neglecting the effect of perceived time pressure may result in an inaccurate forecast of choice probability and overestimate the affectability to the policy changes.

Keywords: activity-travel rescheduling, decision making under uncertainty, mixed heuristic model, perceived time pressure, travel demand management

Procedia PDF Downloads 109

Authors: Zhaofeng Li, Jun Kang Chow, Yu-Hsing Wang

Abstract:

This paper attempts to bridge the soil properties and the mechanical response of soil in the discrete element method (DEM) simulation. The artificial neural network (ANN) was therefore adopted, aiming to reproduce the stress-strain-volumetric response when soil properties are given. 31 biaxial shearing tests with varying soil parameters (e.g., initial void ratio and interparticle friction coefficient) were generated using the DEM simulations. Based on these 45 sets of training data, a three-layer neural network was established which can output the entire stress-strain-volumetric curve during the shearing process from the input soil parameters. Beyond the training data, 2 additional sets of data were generated to examine the validity of the network, and the stress-strain-volumetric curves for both cases were well reproduced using this network. Overall, the ANN was found promising in predicting the soil behavior and reducing repetitive simulation work.

Keywords: artificial neural network, discrete element method, soil properties, stress-strain-volumetric response

Procedia PDF Downloads 393

Authors: Wafa' Slaibi Alsharafat

Abstract:

Cloud computing is a modern approach in network environment. According to increased number of network users and online systems, there is a need to help these systems to be away from unauthorized resource access and detect any attempts for privacy contravention. For that purpose, Intrusion Detection System is an effective security mechanism to detect any attempts of attacks for cloud resources and their information. In this paper, Cloud Intrusion Detection System has been proposed in term of reducing or eliminating any attacks. This model concerns about achieving high detection rate after conducting a set of experiments using benchmarks dataset called KDD'99.

Keywords: IDS, cloud computing, anticipating classifier system, intrusion detection

Procedia PDF Downloads 471

Authors: Vrince Vimal, Madhav J. Nigam

Abstract:

Wireless sensor networks have enticed much of the spotlight from researchers all around the world, owing to its extensive applicability in agricultural, industrial and military fields. Energy conservation node deployment stratagems play a notable role for active implementation of Wireless Sensor Networks. Clustering is the approach in wireless sensor networks which improves energy efficiency in the network. The clustering algorithm needs to have an optimum size and number of clusters, as clustering, if not implemented properly, cannot effectively increase the life of the network. In this paper, an algorithm has been proposed to address connectivity issues with the aim of ensuring the uniform energy consumption of nodes in every part of the network. The results obtained after simulation showed that the proposed algorithm has an edge over existing algorithms in terms of throughput and networks lifetime.

Keywords: Wireless Sensor network (WSN), Random Deployment, Clustering, Isolated Nodes, Networks Lifetime

Procedia PDF Downloads 333

Authors: Tsega Asresa, Getahun Tigistu, Melaku Bayih

Abstract:

Computer vision is a subfield of artificial intelligence that allows computers and systems to extract meaning from digital images and video. It is used in a wide range of fields of study, including self-driving cars, video surveillance, medical diagnosis, manufacturing, law, agriculture, quality control, health care, facial recognition, and military applications. Aromatic medicinal plants are botanical raw materials used in cosmetics, medicines, health foods, essential oils, decoration, cleaning, and other natural health products for therapeutic and Aromatic culinary purposes. These plants and their products not only serve as a valuable source of income for farmers and entrepreneurs but also going to export for valuable foreign currency exchange. In Ethiopia, there is a lack of technologies for the classification and identification of Aromatic medicinal plant parts and disease type cured by aromatic medicinal plants. Farmers, industry personnel, academicians, and pharmacists find it difficult to identify plant parts and disease types cured by plants before ingredient extraction in the laboratory. Manual plant identification is a time-consuming, labor-intensive, and lengthy process. To alleviate these challenges, few studies have been conducted in the area to address these issues. One way to overcome these problems is to develop a deep learning model for efficient identification of Aromatic medicinal plant parts with their corresponding disease type. The objective of the proposed study is to identify the aromatic medicinal plant parts and their disease type classification using computer vision technology. Therefore, this research initiated a model for the classification of aromatic medicinal plant parts and their disease type by exploring computer vision technology. Morphological characteristics are still the most important tools for the identification of plants. Leaves are the most widely used parts of plants besides roots, flowers, fruits, and latex. For this study, the researcher used RGB leaf images with a size of 128x128 x3. In this study, the researchers trained five cutting-edge models: convolutional neural network, Inception V3, Residual Neural Network, Mobile Network, and Visual Geometry Group. Those models were chosen after a comprehensive review of the best-performing models. The 80/20 percentage split is used to evaluate the model, and classification metrics are used to compare models. The pre-trained Inception V3 model outperforms well, with training and validation accuracy of 99.8% and 98.7%, respectively.

Keywords: aromatic medicinal plant, computer vision, convolutional neural network, deep learning, plant classification, residual neural network

Procedia PDF Downloads 180

Authors: Junjie Peng, Jinbao Chen, Shuai Kong, Danxu Liu

Abstract:

On the basis of traditional resource allocation strategies, the usage of resources on physical servers in cloud data center is great uncertain. It will cause waste of resources if the assignment of tasks is not enough. On the contrary, it will cause overload if the assignment of tasks is too much. This is especially obvious when the applications are the same type because of its resource preferences. Considering CPU intensive application is one of the most common types of application in the cloud, we studied the optimization strategy for CPU intensive applications on the same server. We used resource preferences to analyze the case that multiple CPU intensive applications run simultaneously, and put forward a model which can predict the execution time for CPU intensive applications which run simultaneously. Based on the prediction model, we proposed the method to select the appropriate number of applications for a machine. Experiments show that the model can predict the execution time accurately for CPU intensive applications. To improve the execution efficiency of applications, we propose a scheduling model based on priority for CPU intensive applications. Extensive experiments verify the validity of the scheduling model.

Keywords: cloud computing, CPU intensive applications, resource optimization, strategy

Procedia PDF Downloads 274

Authors: Kevin D. Manalo, Jumelyn L. Torres, Noel B. Linsangan

Abstract:

This paper focuses on a wireless myoelectric prosthesis of the upper-limb that uses a Multilayer Perceptron Neural network with back propagation. The algorithm is widely used in pattern recognition. The network can be used to train signals and be able to use it in performing a function on their own based on sample inputs. The paper makes use of the Neural Network in classifying the electromyography signal that is produced by the muscle in the amputee’s skin surface. The gathered data will be passed on through the Classification Stage wirelessly through Zigbee Technology. The signal will be classified and trained to be used in performing the arm positions in the prosthesis. Through programming using Verilog and using a Field Programmable Gate Array (FPGA) with Zigbee, the EMG signals will be acquired and will be used for classification. The classified signal is used to produce the corresponding Hand Movements (Open, Pick, Hold, and Grip) through the Zigbee controller. The data will then be processed through the MLP Neural Network using MATLAB which then be used for the surface myoelectric prosthesis. Z-test will be used to display the output acquired from using the neural network.

Keywords: field programmable gate array, multilayer perceptron neural network, verilog, zigbee

Procedia PDF Downloads 386

Authors: K. Govender, M. Sinclair

Abstract:

This paper demonstrates a proof of concept whereby shorter trips and land use densification can be promoted through an alternative approach to planning and implementation of road infrastructure in the South African context. It briefly discusses how the development of the Compact City concept relies on a combination of promoting shorter trips and densification through a change in focus in road infrastructure provision. The methodology developed in this paper uses a traffic model to test the impact of synthesized deterrence functions on congestion locations in the road network through the assignment of traffic on the study network. The results from this study demonstrate that intelligent planning of road infrastructure can indeed promote reduced urban sprawl, increased residential density and mixed-use areas which are supported by an efficient public transport system; and reduced dependence on the freeway network with a fixed road infrastructure budget. The study has resonance for all cities where urban sprawl is seemingly unstoppable.

Keywords: compact cities, densification, road infrastructure planning, transportation modelling

Procedia PDF Downloads 174

Authors: Mehrnaz Aminifar

Abstract:

The relation between textural parameters and casein network on release of aromatic compounds was investigated over 90-days of ripening. Low DE maltodextrin and WPI were used to modify the textural properties of low fat brined cheese. Hardness, brittleness and compaction of casein network were affected by addition of maltodextrin and WPI. Textural properties and aroma release from cheese texture were affected by interaction of WPI protein-cheese protein and maltodexterin-cheese protein.

Keywords: aroma release, brined cheese, maltodexterin, WPI

Procedia PDF Downloads 350