Search results for: noise robustness

Authors: Gurukiran Veerapur, Nytik Birudavolu, Seetharam U. N., Chandravva Hebbi, R. Praneeth Reddy

Abstract:

This work presents a robust approach for the recognition of characters in Telugu and Kannada, two South Indian scripts with structural similarities in characters. To recognize the characters exhaustive datasets are required, but there are only a few publicly available datasets. As a result, we decided to create a dataset for one language (source language),train the model with it, and then test it with the target language.Telugu is the target language in this work, whereas Kannada is the source language. The suggested method makes use of Canny edge features to increase character identification accuracy on pictures with noise and different lighting. A dataset of 45,150 images containing printed Kannada characters was created. The Nudi software was used to automatically generate printed Kannada characters with different writing styles and variations. Manual labelling was employed to ensure the accuracy of the character labels. The deep learning models like CNN (Convolutional Neural Network) and Visual Attention neural network (VAN) are used to experiment with the dataset. A Visual Attention neural network (VAN) architecture was adopted, incorporating additional channels for Canny edge features as the results obtained were good with this approach. The model's accuracy on the combined Telugu and Kannada test dataset was an outstanding 97.3%. Performance was better with Canny edge characteristics applied than with a model that solely used the original grayscale images. The accuracy of the model was found to be 80.11% for Telugu characters and 98.01% for Kannada words when it was tested with these languages. This model, which makes use of cutting-edge machine learning techniques, shows excellent accuracy when identifying and categorizing characters from these scripts.

Keywords: base characters, modifiers, guninthalu, aksharas, vattakshara, VAN

Procedia PDF Downloads 37

Authors: Gaurav Gupta, Jitendra Mahakud

Abstract:

This study focuses on the significance of internal financing constraints on the determination of corporate fixed investments in the case of Indian manufacturing companies. Financing constraints companies which have less internal fund or retained earnings face more transaction and borrowing costs due to imperfections in the capital market. The period of study is 1999-2000 to 2013-2014 and we consider 618 manufacturing companies for which the continuous data is available throughout the study period. The data is collected from PROWESS data base maintained by Centre for Monitoring Indian Economy Pvt. Ltd. Panel data methods like fixed effect and random effect methods are used for the analysis. The Likelihood Ratio test, Lagrange Multiplier test, and Hausman test results conclude the suitability of the fixed effect model for the estimation. The cash flow and liquidity of the company have been used as the proxies for the internal financial constraints. In accordance with various theories of corporate investments, we consider other firm specific variable like firm age, firm size, profitability, sales and leverage as the control variables in the model. From the econometric analysis, we find internal cash flow and liquidity have the significant and positive impact on the corporate investments. The variables like cost of capital, sales growth and growth opportunities are found to be significantly determining the corporate investments in India, which is consistent with the neoclassical, accelerator and Tobin’s q theory of corporate investment. To check the robustness of results, we divided the sample on the basis of cash flow and liquidity. Firms having cash flow greater than zero are put under one group, and firms with cash flow less than zero are put under another group. Also, the firms are divided on the basis of liquidity following the same approach. We find that the results are robust to both types of companies having positive and negative cash flow and liquidity. The results for other variables are also in the same line as we find for the whole sample. These findings confirm that internal financing constraints play a significant role for determination of corporate investment in India. The findings of this study have the implications for the corporate managers to focus on the projects having higher expected cash inflows to avoid the financing constraints. Apart from that, they should also maintain adequate liquidity to minimize the external financing costs.

Keywords: cash flow, corporate investment, financing constraints, panel data method

Procedia PDF Downloads 229

Authors: Maryam Zarabian, Hector Guzman, Pedro Pereira-Almao, Abraham Fapojuwo

Abstract:

Dry reforming of methane (DRM) has sparked significant industrial and scientific interest not only as a viable alternative for addressing the environmental concerns of two main contributors of the greenhouse effect, i.e., carbon dioxide (CO₂) and methane (CH₄), but also produces syngas, i.e., a mixture of hydrogen (H₂) and carbon monoxide (CO) utilized by a wide range of downstream processes as a feedstock for other chemical productions. In this study, we develop an AI-enable syngas production model to tackle the problem of achieving an equivalent H₂/CO ratio [1:1] with respect to the most efficient conversion. Firstly, the unsupervised density-based spatial clustering of applications with noise (DBSAN) algorithm removes outlier data points from the original experimental dataset. Then, random forest (RF) and deep neural network (DNN) models employ the error-free dataset to predict the DRM results. DNN models inherently would not be able to obtain accurate predictions without a huge dataset. To cope with this limitation, we employ reusing pre-trained layers’ approaches such as transfer learning and greedy layer-wise pretraining. Compared to the other deep models (i.e., pure deep model and transferred deep model), the greedy layer-wise pre-trained deep model provides the most accurate prediction as well as similar accuracy to the RF model with R² values 1.00, 0.999, 0.999, 0.999, 0.999, and 0.999 for the total outlet flow, H₂/CO ratio, H₂ yield, CO yield, CH₄ conversion, and CO₂ conversion outputs, respectively.

Keywords: artificial intelligence, dry reforming of methane, artificial neural network, deep learning, machine learning, transfer learning, greedy layer-wise pretraining

Procedia PDF Downloads 70

Authors: Pius Marthin, Nihal Ata Tutkun

Abstract:

Survival analysis has become one of the paramount procedures in the modeling of time-to-event data. When we encounter complex survival problems, the traditional approach remains limited in accounting for the complex correlational structure between the covariates and the outcome due to the strong assumptions that limit the inference and prediction ability of the resulting models. Several studies exist on the deep learning approach to survival modeling; moreover, the application for the case of complex survival problems still needs to be improved. In addition, the existing models need to address the data structure's complexity fully and are subject to noise and redundant information. In this study, we design a deep learning technique (CmpXRnnSurv_AE) that obliterates the limitations imposed by traditional approaches and addresses the above issues to jointly predict the risk-specific probabilities and survival function for recurrent events with competing risks. We introduce the component termed Risks Information Weights (RIW) as an attention mechanism to compute the weighted cumulative incidence function (WCIF) and an external auto-encoder (ExternalAE) as a feature selector to extract complex characteristics among the set of covariates responsible for the cause-specific events. We train our model using synthetic and real data sets and employ the appropriate metrics for complex survival models for evaluation. As benchmarks, we selected both traditional and machine learning models and our model demonstrates better performance across all datasets.

Keywords: cumulative incidence function (CIF), risk information weight (RIW), autoencoders (AE), survival analysis, recurrent events with competing risks, recurrent neural networks (RNN), long short-term memory (LSTM), self-attention, multilayers perceptrons (MLPs)

Procedia PDF Downloads 72

Authors: Amir Abbas Fatemi, Zahra Tabrizian, Kabir Sadeghi

Abstract:

To find the location and severity of damage that occurs in a structure, characteristics changes in dynamic and static can be used. The non-destructive techniques are more common, economic, and reliable to detect the global or local damages in structures. This paper presents a non-destructive method in structural damage detection and assessment using GA and static data. Thus, a set of static forces is applied to some of degrees of freedom and the static responses (displacements) are measured at another set of DOFs. An analytical model of the truss structure is developed based on the available specification and the properties derived from static data. The damages in structure produce changes to its stiffness so this method used to determine damage based on change in the structural stiffness parameter. Changes in the static response which structural damage caused choose to produce some simultaneous equations. Genetic Algorithms are powerful tools for solving large optimization problems. Optimization is considered to minimize objective function involve difference between the static load vector of damaged and healthy structure. Several scenarios defined for damage detection (single scenario and multiple scenarios). The static damage identification methods have many advantages, but some difficulties still exist. So it is important to achieve the best damage identification and if the best result is obtained it means that the method is Reliable. This strategy is applied to a plane truss. This method is used for a plane truss. Numerical results demonstrate the ability of this method in detecting damage in given structures. Also figures show damage detections in multiple damage scenarios have really efficient answer. Even existence of noise in the measurements doesn’t reduce the accuracy of damage detections method in these structures.

Keywords: damage detection, finite element method, static data, non-destructive, genetic algorithm

Procedia PDF Downloads 214

Authors: Davies Obaromi, Qin Yongsong, James Ndege

Abstract:

To interpolate scattered or regularly distributed data, there are imprecise or exact methods. However, there are some of these methods that could be used for interpolating data in a regular grid and others in an irregular grid. In spatial epidemiology, it is important to examine how a disease prevalence rates are distributed in space, and how they relate with each other within a defined distance and direction. In this study, for the geographic and graphic representation of the disease prevalence, linear and biharmonic spline methods were implemented in MATLAB, and used to identify, localize and compare for smoothing in the distribution patterns of tuberculosis (TB) in Eastern Cape Province. The aim of this study is to produce a more “smooth” graphical disease map for TB prevalence patterns by a 3-D curve fitting techniques, especially the biharmonic splines that can suppress noise easily, by seeking a least-squares fit rather than exact interpolation. The datasets are represented generally as a 3D or XYZ triplets, where X and Y are the spatial coordinates and Z is the variable of interest and in this case, TB counts in the province. This smoothing spline is a method of fitting a smooth curve to a set of noisy observations using a spline function, and it has also become the conventional method for its high precision, simplicity and flexibility. Surface and contour plots are produced for the TB prevalence at the provincial level for 2012 – 2015. From the results, the general outlook of all the fittings showed a systematic pattern in the distribution of TB cases in the province and this is consistent with some spatial statistical analyses carried out in the province. This new method is rarely used in disease mapping applications, but it has a superior advantage to be assessed at subjective locations rather than only on a rectangular grid as seen in most traditional GIS methods of geospatial analyses.

Keywords: linear, biharmonic splines, tuberculosis, South Africa

Procedia PDF Downloads 227

Authors: Si-Geun Choi, Hoon-Jae Park, Jung-Woo Cho, Jin-Ho Lim, Jin-Young Park, Joo-Young Oh, Jae-Il Jeong Seock-Sam Kim, Young Tae Cho, Chan Gyu Kim, Jong-Hyoung Kim

Abstract:

The tilting pad bearing is a kind of the fluid film bearing and it can contribute to the high speed and the high load performance compared to other bearings including the rolling element bearing. Furthermore, the tilting bearing has many advantages such as high stability at high-speed performance, long life, high damping, high impact resistance and low noise. Therefore, it mostly used in mid to large size turbomachines, despite the high price disadvantage. Recently, manufacture and process employing laser techniques advancing at a fast-growing rate in mechanical industry, the dissimilar metal weld process employing laser techniques is actively studied. Moreover, also, Industry fields try to apply for welding the white metal and the back metal using laser cladding method for high durability. Furthermore, it has followed that laser cladding method has a lot better bond strength, toughness, anti-abrasion and environment-friendly than centrifugal casting method through preceding research. Therefore, the laser cladding method has a lot better quality, cost reduction, eco-friendliness and permanence of technology than the centrifugal casting method or the gravity casting method. In this study, we compare the mechanical properties of different bearing materials by evaluating the behavior of laser cladding layer with various materials (i.e. SS400, SCM440, S20C) under the same parameters. Furthermore, we analyze the porosity of various tilting pad bearing materials which white metal treated on samples. SEM, EDS analysis and hardness tests of three materials are shown to understand the mechanical properties and tribological behavior. W/D ratio, surface roughness results with various materials are performed in this study.

Keywords: laser cladding, tilting pad bearing, white metal, mechanical properties

Procedia PDF Downloads 364

Authors: Sujata Kulkarni, Udhav Bhosle, Vijaykumar T.

Abstract:

Due to the closeness between seismic signals and non-seismic signals, it is vital to detect earthquakes using conventional methods. In order to distinguish between seismic events and non-seismic events depending on their amplitude, our study processes the data that come from seismic sensors. The authors suggest a robust noise suppression technique that makes use of a bandpass filter, an IIR Wiener filter, recursive short-term average/long-term average (STA/LTA), and Carl short-term average (STA)/long-term average for event identification (LTA). The trigger ratio used in the proposed study to differentiate between seismic and non-seismic activity is determined. The proposed work focuses on significant feature extraction for machine learning-based seismic event detection. This serves as motivation for compiling a dataset of all features for the identification and forecasting of seismic signals. We place a focus on feature vector dimension reduction techniques due to the temporal complexity. The proposed notable features were experimentally tested using a machine learning model, and the results on unseen data are optimal. Finally, a presentation using a hybrid dataset (captured by different sensors) demonstrates how this model may also be employed in a real-time setting while lowering false alarm rates. The planned study is based on the examination of seismic signals obtained from both individual sensors and sensor networks (SN). A wideband seismic signal from BSVK and CUKG station sensors, respectively located near Basavakalyan, Karnataka, and the Central University of Karnataka, makes up the experimental dataset.

Keywords: Carl STA/LTA, features extraction, real time, dataset, machine learning, seismic detection

Procedia PDF Downloads 100

Authors: N. Benkaci, E. Oubaiche, J.-L. Chatelain, R. Bensalem, K. Abbes

Abstract:

The ambient vibration H/V technique is widely used nowadays in microzonation studies, because of its easy field handling and its low cost, compared to other geophysical methods. However, in presence of complex geology or lateral heterogeneity evidenced by more than one peak frequency in the H/V curve, it is difficult to interpret the results, especially when soil information is lacking. In this work, we focus on the construction site of the Baraki 40000=place stadium, located in the north-east side of the Mitidja basin (Algeria), to identify the seismic wave amplification zones. H/V curve analysis leads to the observation of spatial and time variability of the H/V frequency peaks. The spatial variability allows dividing the studied area into three main zones: (1) one with a predominant frequency around 1,5 Hz showing an important amplification level, (2) the second exhibits two peaks at 1,5 Hz and in the 4 Hz – 10 Hz range, and (3) the third zone is characterized by a plateau between 2 Hz and 3 Hz. These H/V curve categories reveal a consequent lateral heterogeneity dividing the stadium site roughly in the middle. Furthermore, a continuous ambient vibration recording during several weeks allows showing that the first peak at 1,5 Hz in the second zone, completely disappears between 2 am and 4 am, and reaching its maximum amplitude around 12 am. Consequently, the anthropogenic noise source generating these important variations could be the Algiers Rocade Sud highway, located in the maximum amplification azimuth direction of the H/V curves. This work points out that the H/V method is an important tool to perform nano-zonation studies prior to geotechnical and geophysical investigations, and that, in some cases, the H/V technique fails to reveal the resonance frequency in the absence of strong anthropogenic source.

Keywords: ambient vibrations, amplification, fundamental frequency, lateral heterogeneity, site effect

Procedia PDF Downloads 229

Authors: Seyedamir Makinejadsanij

Abstract:

One of the most important factors in the production of quality steel is to know the exact weight of steel in the steelmaking area. In this study, a calculation method is presented to estimate the exact weight of the melt as well as the objects transported by the overhead crane. Iran Alloy Steel Company's steelmaking area has three 90-ton cranes, which are responsible for transferring the ladles and ladle caps between 34 areas in the melt shop. Each crane is equipped with a Disomat Tersus weighing system that calculates and displays real-time weight. The moving object has a variable weight due to swinging, and the weighing system has an error of about +-5%. This means that when the object is moving by a crane, which weighs about 80 tons, the device (Disomat Tersus system) calculates about 4 tons more or 4 tons less, and this is the biggest problem in calculating a real weight. The k-means algorithm is an unsupervised clustering method that was used here. The best result was obtained by considering 3 centers. Compared to the normal average(one) or two, four, five, and six centers, the best answer is with 3 centers, which is logically due to the elimination of noise above and below the real weight. Every day, the standard weight is moved with working cranes to test and calibrate cranes. The results are shown that the accuracy is about 40 kilos per 60 tons (standard weight). As a result, with this method, the accuracy of moving weight is calculated as 99.95%. K-means is used to calculate the exact mean of objects. The stopping criterion of the algorithm is also the number of 1000 repetitions or not moving the points between the clusters. As a result of the implementation of this system, the crane operator does not stop while moving objects and continues his activity regardless of weight calculations. Also, production speed increased, and human error decreased.

Keywords: k-means, overhead crane, melt weight, weight estimation, swing problem

Procedia PDF Downloads 79

Authors: Yuekun Chen, Yousef Sardahi, Salam Hajjar, Christopher Greer

Abstract:

This paper presents a multi-objective optimal design of a cascade control system for an underactuated mechanical system. Cascade control structures usually include two control algorithms (inner and outer). To design such a control system properly, the following conflicting objectives should be considered at the same time: 1) the inner closed-loop control must be faster than the outer one, 2) the inner loop should fast reject any disturbance and prevent it from propagating to the outer loop, 3) the controlled system should be insensitive to measurement noise, and 4) the controlled system should be driven by optimal energy. Such a control problem can be formulated as a multi-objective optimization problem such that the optimal trade-offs among these design goals are found. To authors best knowledge, such a problem has not been studied in multi-objective settings so far. In this work, an underactuated mechanical system consisting of a rotary servo motor and a ball and beam is used for the computer simulations, the setup parameters of the inner and outer control systems are tuned by NSGA-II (Non-dominated Sorting Genetic Algorithm), and the dominancy concept is used to find the optimal design points. The solution of this problem is not a single optimal cascade control, but rather a set of optimal cascade controllers (called Pareto set) which represent the optimal trade-offs among the selected design criteria. The function evaluation of the Pareto set is called the Pareto front. The solution set is introduced to the decision-maker who can choose any point to implement. The simulation results in terms of Pareto front and time responses to external signals show the competing nature among the design objectives. The presented study may become the basis for multi-objective optimal design of multi-loop control systems.

Keywords: cascade control, multi-Loop control systems, multiobjective optimization, optimal control

Procedia PDF Downloads 138

Authors: Hussain Sajid, Jung-Hun Shin, Kum-Won Cho

Abstract:

Image super-resolution is the most common computer vision problem with many important applications. Generative adversarial networks (GANs) have promoted remarkable advances in single-image super-resolution (SR) by recovering photo-realistic images. However, high memory requirements of GAN-based SR (mainly generators) lead to performance degradation and increased energy consumption, making it difficult to implement it onto resource-constricted devices. To relieve such a problem, In this paper, we introduce an optimized and highly efficient architecture for SR-GAN (generator) model by utilizing model compression techniques such as Knowledge Distillation and pruning, which work together to reduce the storage requirement of the model also increase in their performance. Our method begins with distilling the knowledge from a large pre-trained model to a lightweight model using different loss functions. Then, iterative weight pruning is applied to the distilled model to remove less significant weights based on their magnitude, resulting in a sparser network. Knowledge Distillation reduces the model size by 40%; pruning then reduces it further by 18%. To accelerate the learning process, we employ the Horovod framework for distributed training on a cluster of 2 nodes, each with 8 GPUs, resulting in improved training performance and faster convergence. Experimental results on various benchmarks demonstrate that the proposed compressed model significantly outperforms state-of-the-art methods in terms of peak signal-to-noise ratio (PSNR), structural similarity index measure (SSIM), and image quality for x4 super-resolution tasks.

Keywords: single-image super-resolution, generative adversarial networks, knowledge distillation, pruning

Procedia PDF Downloads 70

Authors: J. Alvarez, J. Pimienta, R. Sarmiento

Abstract:

Diode lasers working in free mode present different shifting and broadening determined by external factors such as temperature, current or mechanical vibrations, and they are not more useful in applications such as spectroscopy, metrology, and cooling of atoms, among others. Different configurations can reduce the spectral width of a laser; one of the most effective is to extend the optical resonator of the laser diode and use optical feedback either with the help of a partially reflective mirror or with a diffraction grating; this latter configuration is not only allowed to reduce the spectral width of the laser line but also to coarsely adjust its working wavelength, within a wide range typically ~ 10nm by slightly varying the angle of the diffraction grating. Two settings are commonly used for this purpose, the Littrow configuration and the Littmann Metcalf. In this paper, we present the design, construction, and characterization of a compact extended laser cavity in Littrow configuration. The designed cavity is compact and was machined on an aluminum block using computer numerical control (CNC); it has a mass of only 380 g. The design was tested on laser diodes with different wavelengths, 650nm, 780nm, and 795 nm, but can be equally efficient at other wavelengths. This report details the results obtained from the extended cavity working at a wavelength of 780 nm, with an output power of around 35mW and a line width of less than 1Mhz. The cavity was used to observe the spectrum of the corresponding Rubidium D2 line. By modulating the current and with the help of phase detection techniques, a dispersion signal with an excellent signal-to-noise ratio was generated that allowed the stabilization of the laser to a transition of the hyperfine structure of Rubidium with an integral proportional controller (PI) circuit made with precision operational amplifiers.

Keywords: Littrow, Littman-Metcalf, line width, laser stabilization, hyperfine structure

Procedia PDF Downloads 211

Authors: Swathi Gopakumar, Sruthi Krishna, Shivasubramani Krishnamoorthy

Abstract:

Contactless, painless and radiation-free thermal imaging technology is one of the preferred screening modalities for detection of breast cancer. However, poor signal to noise ratio and the inexorable need to preserve edges defining cancer cells and normal cells, make the segmentation process difficult and hence unsuitable for computer-aided diagnosis of breast cancer. This paper presents key findings from a research conducted on the appraisal of two promising techniques, for the detection of breast cancer: (I) marker-controlled, Level-set segmentation of anisotropic diffusion filtered preprocessed image versus (II) Segmentation using marker-controlled level-set on a Gaussian-filtered image. Gaussian-filtering processes the image uniformly, whereas anisotropic filtering processes only in specific areas of a thermographic image. The pre-processed (Gaussian-filtered and anisotropic-filtered) images of breast samples were then applied for segmentation. The segmentation of breast starts with initial level-set function. In this study, marker refers to the position of the image to which initial level-set function is applied. The markers are generally placed on the left and right side of the breast, which may vary with the breast size. The proposed method was carried out on images from an online database with samples collected from women of varying breast characteristics. It was observed that the breast was able to be segmented out from the background by adjustment of the markers. From the results, it was observed that as a pre-processing technique, anisotropic filtering with level-set segmentation, preserved the edges more effectively than Gaussian filtering. Segmented image, by application of anisotropic filtering was found to be more suitable for feature extraction, enabling automated computer-aided diagnosis of breast cancer.

Keywords: anisotropic diffusion, breast, Gaussian, level-set, thermograms

Procedia PDF Downloads 363

Authors: Ballout Amor, Yasmina Bouchahm, Lacheheb Dhia Eddine Zakaria

Abstract:

Several studies all around the world were conducted on the phenomenon of the urban heat island, and referring to the results obtained, one of the most important factors that influence this phenomenon is the mineralization of the cities which means the reducing of evaporative urban surfaces, replacing vegetation and wetlands with concrete and asphalt. The use of vegetation and water can change the urban environment and improve comfort, thus reduce the heat island. The trees act as a mask to the sun, wind, and sound, and also as a source of humidity which reduces air temperature and surrounding surfaces. Water also acts as a buffer to noise; it is also a source of moisture and regulates temperature not to mention the psychological effect on humans. Our main objective in this paper is to determine the impact of vegetation, ponds and fountains on the urban micro climate in general and on the thermal comfort of people along the Independence square in the Algerian city of Sétif, which is a semi-arid climate, in particularly. In order to reach this objective, a comparative study between different scenarios has been done; the use of the Envi-met program enabled us to model the urban environment of the Independence Square and to study the possibility of improving the conditions of comfort by adding an amount of vegetation and water ponds. After studying the results obtained (temperature, relative humidity, wind speed, PMV and PPD indicators), the efficiency of the additions we've made on the square was confirmed and this is what helped us to confirm our assumptions regarding the terms of comfort in the studied site, and in the end we are trying to develop recommendations and solutions which may contribute to improve the conditions for greater comfort in the Independence square.

Keywords: comfort in outer space, urban environment, scenarisation, vegetation, water ponds, public square, simulation

Procedia PDF Downloads 438

Authors: Penina Orenstein

Abstract:

In today’s world, particularly with COVID-19 a constant worldwide threat, organizations need greater visibility over their supply chains more than ever before, in order to find areas for improvement and greater efficiency, reduce the chances of disruption and stay competitive. The concept of supply chain mapping is one where every process and route is mapped in detail between each vendor and supplier. The simplest method of mapping involves sourcing publicly available data including news and financial information concerning relationships between suppliers. An additional layer of information would be disclosed by large, direct suppliers about their production and logistics sites. While this method has the advantage of not requiring any input from suppliers, it also doesn’t allow for much transparency beyond the first supplier tier and may generate irrelevant data—noise—that must be filtered out to find the actionable data. The primary goal of this research is to build data maps of supply chains by focusing on a layered approach. Using these maps, the secondary goal is to address the question as to whether the supply chain is re-engineered to make improvements, for example, to lower the carbon footprint. Using a drill-down approach, the end result is a comprehensive map detailing the linkages between tier-one, tier-two, and tier-three suppliers super-imposed on a geographical map. The driving force behind this idea is to be able to trace individual parts to the exact site where they’re manufactured. In this way, companies can ensure sustainability practices from the production of raw materials through the finished goods. The approach allows companies to identify and anticipate vulnerabilities in their supply chain. It unlocks predictive analytics capabilities and enables them to act proactively. The research is particularly compelling because it unites network science theory with empirical data and presents the results in a visual, intuitive manner.

Keywords: data mining, supply chain, empirical research, data mapping

Procedia PDF Downloads 158

Authors: Arindam Chaudhuri

Abstract:

Classification of data has been actively used for most effective and efficient means of conveying knowledge and information to users. The prima face has always been upon techniques for extracting useful knowledge from data such that returns are maximized. With emergence of huge datasets the existing classification techniques often fail to produce desirable results. The challenge lies in analyzing and understanding characteristics of massive data sets by retrieving useful geometric and statistical patterns. We propose a supervised parallel fuzzy rough support vector machine (PFRSVM) for data classification in cloud environment. The classification is performed by PFRSVM using hyperbolic tangent kernel. The fuzzy rough set model takes care of sensitiveness of noisy samples and handles impreciseness in training samples bringing robustness to results. The membership function is function of center and radius of each class in feature space and is represented with kernel. It plays an important role towards sampling the decision surface. The success of PFRSVM is governed by choosing appropriate parameter values. The training samples are either linear or nonlinear separable. The different input points make unique contributions to decision surface. The algorithm is parallelized with a view to reduce training times. The system is built on support vector machine library using Hadoop implementation of MapReduce. The algorithm is tested on large data sets to check its feasibility and convergence. The performance of classifier is also assessed in terms of number of support vectors. The challenges encountered towards implementing big data classification in machine learning frameworks are also discussed. The experiments are done on the cloud environment available at University of Technology and Management, India. The results are illustrated for Gaussian RBF and Bayesian kernels. The effect of variability in prediction and generalization of PFRSVM is examined with respect to values of parameter C. It effectively resolves outliers’ effects, imbalance and overlapping class problems, normalizes to unseen data and relaxes dependency between features and labels. The average classification accuracy for PFRSVM is better than other classifiers for both Gaussian RBF and Bayesian kernels. The experimental results on both synthetic and real data sets clearly demonstrate the superiority of the proposed technique.

Keywords: FRSVM, Hadoop, MapReduce, PFRSVM

Procedia PDF Downloads 477

Authors: Narendra Akhadkar, Silvestre Cano, Christophe Gourru

Abstract:

Injection-molded parts are widely used in power system protection products. One of the biggest challenges in an injection molding process is shrinkage and warpage of the molded parts. All these geometrical variations may have an adverse effect on the quality of the product, functionality, cost, and time-to-market. The situation becomes more challenging in the case of intricate shapes and in mass production using multi-cavity tools. To control the effects of shrinkage and warpage, it is very important to correctly find out the input parameters that could affect the product performance. With the advances in the computer-aided engineering (CAE), different tools are available to simulate the injection molding process. For our case study, we used the MoldFlow insight tool. Our aim is to predict the spread of the functional dimensions and geometrical variations on the part due to variations in the input parameters such as material viscosity, packing pressure, mold temperature, melt temperature, and injection speed. The input parameters may vary during batch production or due to variations in the machine process settings. To perform the accurate product assembly variation simulation, the first step is to perform an individual part variation simulation to render realistic tolerance ranges. In this article, we present a method to simulate part variations coming from the input parameters variation during batch production. The method is based on computer simulations and experimental validation using the full factorial design of experiments (DoE). The robustness of the simulation model is verified through input parameter wise sensitivity analysis study performed using simulations and experiments; all the results show a very good correlation in the material flow direction. There exists a non-linear interaction between material and the input process variables. It is observed that the parameters such as packing pressure, material, and mold temperature play an important role in spread on functional dimensions and geometrical variations. This method will allow us in the future to develop accurate/realistic virtual prototypes based on trusted simulated process variation and, therefore, increase the product quality and potentially decrease the time to market.

Keywords: correlation, molding process, tolerance, sensitivity analysis, variation simulation

Procedia PDF Downloads 158

Authors: Gulfam Haider, sana danish

Abstract:

Convolutional Neural Networks (CNNs) have emerged as powerful tools for image classification, and the choice of optimizers profoundly affects their performance. The study of optimizers and their adaptations remains a topic of significant importance in machine learning research. While numerous studies have explored and advocated for various optimizers, the efficacy of these optimization techniques is still subject to scrutiny. This work aims to address the challenges surrounding the effectiveness of optimizers by conducting a comprehensive analysis and evaluation. The primary focus of this investigation lies in examining the performance of different optimizers when employed in conjunction with the popular activation function, Rectified Linear Unit (ReLU). By incorporating ReLU, known for its favorable properties in prior research, the aim is to bolster the effectiveness of the optimizers under scrutiny. Specifically, we evaluate the adjustment of these optimizers with both the original Softmax activation function and the modified ReLU activation function, carefully assessing their impact on overall performance. To achieve this, a series of experiments are conducted using a well-established benchmark dataset for image classification tasks, namely the Canadian Institute for Advanced Research dataset (CIFAR-10). The selected optimizers for investigation encompass a range of prominent algorithms, including Adam, Root Mean Squared Propagation (RMSprop), Adaptive Learning Rate Method (Adadelta), Adaptive Gradient Algorithm (Adagrad), and Stochastic Gradient Descent (SGD). The performance analysis encompasses a comprehensive evaluation of the classification accuracy, convergence speed, and robustness of the CNN models trained with each optimizer. Through rigorous experimentation and meticulous assessment, we discern the strengths and weaknesses of the different optimization techniques, providing valuable insights into their suitability for image classification tasks. By conducting this in-depth study, we contribute to the existing body of knowledge surrounding optimizers in CNNs, shedding light on their performance characteristics for image classification. The findings gleaned from this research serve to guide researchers and practitioners in making informed decisions when selecting optimizers and activation functions, thus advancing the state-of-the-art in the field of image classification with convolutional neural networks.

Keywords: deep neural network, optimizers, RMsprop, ReLU, stochastic gradient descent

Procedia PDF Downloads 102

Authors: T. Elrazaz

Abstract:

This paper examines accrual and real earnings management by publicly listed targets around mergers and acquisitions. Prior literature shows that earnings management around mergers and acquisitions can have a significant economic impact because of the associated wealth transfers among stakeholders. More importantly, acting on behalf of their shareholders or pursuing their self-interests, managers of both targets and acquirers may be equally motivated to manipulate earnings prior to an acquisition to generate higher gains for their shareholders or themselves. Building on the grounds of information asymmetry, agency conflicts, stewardship theory, and the revelation principle, this study addresses the question of whether takeover targets employ accrual and real earnings management in the periods prior to the announcement of Mergers and Acquisitions (M&A). Additionally, this study examines whether acquirers are able to detect targets’ earnings management, and in response, adjust the acquisition premium paid in order not to face the risk of overpayment. This study uses an aggregate accruals approach in estimating accrual earnings management as proxied by estimated abnormal accruals. Additionally, real earnings management is proxied for by employing widely used models in accounting and finance literature. The results of this study indicate that takeover targets manipulate their earnings using accruals in the second year with an earnings release prior to the announcement of the M&A. Moreover, in partitioning the sample of targets according to the method of payment used in the deal, the results are restricted only to targets of stock-financed deals. These results are consistent with the argument that targets of cash-only or mixed-payment deals do not have the same strong motivations to manage their earnings as their stock-financed deals counterparts do additionally supporting the findings of prior studies that the method of payment in takeovers is value relevant. The findings of this study also indicate that takeover targets manipulate earnings upwards through cutting discretionary expenses the year prior to the acquisition while they do not do so by manipulating sales or production costs. Moreover, in partitioning the sample of targets according to the method of payment used in the deal, the results are restricted only to targets of stock-financed deals, providing further robustness to the results derived under the accrual-based models. Finally, this study finds evidence suggesting that acquirers are fully aware of the accrual-based techniques employed by takeover targets and can unveil such manipulation practices. These results are robust to alternative accrual and real earnings management proxies, as well as controlling for the method of payment in the deal.

Keywords: accrual earnings management, acquisition premium, real earnings management, takeover targets

Procedia PDF Downloads 98

Authors: Khalil Ahmad Kakar

Abstract:

In developing countries increasing the automobile and low capacity public transport (para-transit) which are creating congestion, pollution, noise, and traffic accident are the most critical quandary. These issues are under the analysis of assessors to break down the puzzle and propose sustainable urban public transport system. Kabul city is one of those urban areas that the inhabitants are suffering from lack of tolerable and friendly public transport system. The city is the most-populous and overcrowded with around 4.5 million population. The para-transit is the only dominant public transit system with a very poor level of services and low capacity vehicles (6-20 passengers). Therefore, this study after detailed investigations suggests bus rapid transit (BRT) system in Kabul City. It is aimed to mitigate the role of informal transport and decreases congestion. The research covers three parts. In the first part, aggregated travel demand modelling (four-step) is applied to determine the number of users for para-transit and assesses BRT network based on higher passenger demand for public transport mode. In the second part, state preference (SP) survey and binary logit model are exerted to figure out the utility of existing para-transit mode and planned BRT system. Finally, the impact of predicted BRT system on para-transit is evaluated. The extracted outcome based on high travel demand suggests 10 km network for the proposed BRT system, which is originated from the district tenth and it is ended at Kabul International Airport. As well as, the result from the disaggregate travel mode-choice model, based on SP and logit model indicates that the predicted mass rapid transit system has higher utility with the significant impact regarding the reduction of para-transit.

Keywords: BRT, para-transit, travel demand modelling, Kabul City, logit model

Procedia PDF Downloads 169

Authors: Hachaichi Mohamed Nour El-Islem, Baouni Tahar

Abstract:

Environmental sustainability rather than a trans-disciplinary and a scientific issue, is the main problem that characterizes all modern cities nowadays. In developing countries, this concern is expressed in a plethora of critical urban ills: traffic congestion, air pollution, noise, urban decay, increase in energy consumption and CO₂ emissions which blemish cities’ landscape and might threaten citizens’ health and welfare. As in the same manner as developing world cities, the rapid growth of Algiers’ human population and increasing in city scale phenomena lead eventually to increase in daily trips, energy consumption and CO₂ emissions. In addition, the lack of proper and sustainable planning of the city’s infrastructure is one of the most relevant issues from which Algiers suffers. The aim of this contribution is to estimate the carbon deficit of the City of Algiers, Algeria, using the Ecological Footprint Model (carbon footprint). In order to achieve this goal, the amount of CO₂ from fuel combustion has been calculated and aggregated into five sectors (agriculture, industry, residential, tertiary and transportation); as well, Algiers’ biocapacity (CO₂ uptake land) has been calculated to determine the ecological overshoot. This study shows that Algiers’ transport system is not sustainable and is generating more than 50% of Algiers total carbon footprint which cannot be sequestered by the local forest land. The aim of this research is to show that the Carbon Footprint Assessment might be a relevant indicator to design sustainable strategies/policies striving to reduce CO₂ by setting in motion the energy consumption in the transportation sector and reducing the use of fossil fuels as the main energy input.

Keywords: biocapacity, carbon footprint, ecological footprint assessment, energy consumption

Procedia PDF Downloads 134

Authors: Ainouna Bouziane

Abstract:

The ability of Electron Tomography to recover the 3D structure of catalysts, with spatial resolution in the subnanometer scale, has been widely explored and reviewed in the last decades. A variety of experimental techniques, based either on Transmission Electron Microscopy (TEM) or Scanning Transmission Electron Microscopy (STEM) have been used to reveal different features of nanostructured catalysts in 3D, but High Angle Annular Dark Field imaging in STEM mode (HAADF-STEM) stands out as the most frequently used, given its chemical sensitivity and avoidance of imaging artifacts related to diffraction phenomena when dealing with crystalline materials. In this regard, our group has developed a methodology that combines image denoising by undecimated wavelet transforms (UWT) with automated, advanced segmentation procedures and parameter selection methods using CS-TVM (Compressed Sensing-total variation minimization) algorithms to reveal more reliable quantitative information out of the 3D characterization studies. However, evaluating the accuracy of the magnitudes estimated from the segmented volumes is also an important issue that has not been properly addressed yet, because a perfectly known reference is needed. The problem particularly complicates in the case of multicomponent material systems. To tackle this key question, we have developed a methodology that incorporates volume reconstruction/segmentation methods. In particular, we have established an approach to evaluate, in quantitative terms, the accuracy of TVM reconstructions, which considers the influence of relevant experimental parameters like the range of tilt angles, image noise level or object orientation. The approach is based on the analysis of material-realistic, 3D phantoms, which include the most relevant features of the system under analysis.

Keywords: electron tomography, supported catalysts, nanometrology, error assessment

Procedia PDF Downloads 65

Authors: Amgad A. Farghal, Dina I. El Desouki

Abstract:

Energy consumption and users’ satisfaction were compared in three LEED certified office buildings in turkey and an office building in Egypt. The field studies were conducted in summer 2012. The measured environmental parameters in the four buildings were indoor air temperature, relative humidity, CO2 percentage and light intensity. The traditional building is located in Smart Village in Abu Rawash, Cairo, Egypt. The building was studied for 7 days resulting in 84 responds. The three rated buildings are in Istanbul; Turkey. A Platinum LEED certified office building is owned by BASF and gained a platinum certificate for new construction and major renovation. The building was studied for 3 days resulting in 13 responds. A Gold LEED certified office building is owned by BASF and gained a gold certificate for new construction and major renovation. The building was studied for 2 days resulting in 10 responds. A silver LEED certified office building is owned by Unilever and gained a silver certificate for commercial interiors. The building was studied for 7 days resulting in 84 responds. The results showed that all buildings had no significant difference regarding occupants’ satisfaction with the amount of lighting, noise level, odor and access to the outdoor view. There was significant difference between occupants’ satisfaction in LEED certified buildings and the traditional building regarding the thermal environment and the perception of the general environment (colors, carpet and decoration. The findings suggest that careful design could lead to a certified building that enhances the thermal environment and the perception of the indoor environment leading to energy consumption without scarifying occupants’ satisfaction.

Keywords: energy consumption, occupants’ satisfaction, rating systems, office buildings

Procedia PDF Downloads 404

Authors: V. Veeraprathap, G. S. Harish, G. Narendra Kumar

Abstract:

Uncontrolled growth of abnormal cells in the lung in the form of tumor can be either benign (non-cancerous) or malignant (cancerous). Patients with Lung Cancer (LC) have an average of five years life span expectancy provided diagnosis, detection and prediction, which reduces many treatment options to risk of invasive surgery increasing survival rate. Computed Tomography (CT), Positron Emission Tomography (PET), and Magnetic Resonance Imaging (MRI) for earlier detection of cancer are common. Gaussian filter along with median filter used for smoothing and noise removal, Histogram Equalization (HE) for image enhancement gives the best results without inviting further opinions. Lung cavities are extracted and the background portion other than two lung cavities is completely removed with right and left lungs segmented separately. Region properties measurements area, perimeter, diameter, centroid and eccentricity measured for the tumor segmented image, while texture is characterized by Gray-Level Co-occurrence Matrix (GLCM) functions, feature extraction provides Region of Interest (ROI) given as input to classifier. Two levels of classifications, K-Nearest Neighbor (KNN) is used for determining patient condition as normal or abnormal, while Artificial Neural Networks (ANN) is used for identifying the cancer stage is employed. Discrete Wavelet Transform (DWT) algorithm is used for the main feature extraction leading to best efficiency. The developed technology finds encouraging results for real time information and on line detection for future research.

Keywords: artificial neural networks, ANN, discrete wavelet transform, DWT, gray-level co-occurrence matrix, GLCM, k-nearest neighbor, KNN, region of interest, ROI

Procedia PDF Downloads 131

Authors: Esmat Kamel

Abstract:

This paper aims to shed light on the extent to which the Agadir Association agreement has fostered inter regional trade between the E.U_26 and the Agadir_4 countries; once that we control for the evolution of Agadir agreement’s exports to the rest of the world. The next valid question will be regarding any remarkable variation in the spatial/sectoral structure of exports, and to what extent has it been induced by the Agadir agreement itself and precisely after the adoption of rules of origin and the PANEURO diagonal cumulative scheme? The paper’s empirical dataset covering a timeframe from [2000 -2009] was designed to account for sector specific export and intermediate flows and the bilateral structured gravity model was custom tailored to capture sector and regime specific rules of origin and the Poisson Pseudo Maximum Likelihood Estimator was used to calculate the gravity equation. The methodological approach of this work is considered to be a threefold one which starts first by conducting a ‘Hierarchal Cluster Analysis’ to classify final export flows showing a certain degree of linkage between each other. The analysis resulted in three main sectoral clusters of exports between Agadir_4 and E.U_26: cluster 1 for Petrochemical related sectors, cluster 2 durable goods and finally cluster 3 for heavy duty machinery and spare parts sectors. Second step continues by taking export flows resulting from the 3 clusters to be subject to treatment with diagonal Rules of origin through ‘The Double Differences Approach’, versus an equally comparable untreated control group. Third step is to verify results through a robustness check applied by ‘Propensity Score Matching’ to validate that the same sectoral final export and intermediate flows increased when rules of origin were relaxed. Through all the previous analysis, a remarkable and partial significance of the interaction term combining both treatment effects and time for the coefficients of 13 out of the 17 covered sectors turned out to be partially significant and it further asserted that treatment with diagonal rules of origin contributed in increasing Agadir’s_4 final and intermediate exports to the E.U._26 on average by 335% and in changing Agadir_4 exports structure and composition to the E.U._26 countries.

Keywords: agadir association agreement, structured gravity model, hierarchal cluster analysis, double differences estimation, propensity score matching, diagonal and relaxed rules of origin

Procedia PDF Downloads 307

Authors: Byomakesh Debata, Jitendra Mahakud

Abstract:

The recent financial crisis has been characterized by massive monetary policy interventions by the Central bank, and it has amplified the importance of liquidity for the stability of the stock market. This paper empirically elucidates the actual impact of monetary policy interventions on stock market liquidity covering all National Stock Exchange (NSE) Stocks, which have been traded continuously from 2002 to 2015. The present study employs a multivariate VAR model along with VAR-granger causality test, impulse response functions, block exogeneity test, and variance decomposition to analyze the direction as well as the magnitude of the relationship between monetary policy and market liquidity. Our analysis posits a unidirectional relationship between monetary policy (call money rate, base money growth rate) and aggregate market liquidity (traded value, turnover ratio, Amihud illiquidity ratio, turnover price impact, high-low spread). The impulse response function analysis clearly depicts the influence of monetary policy on stock liquidity for every unit innovation in monetary policy variables. Our results suggest that an expansionary monetary policy increases aggregate stock market liquidity and the reverse is documented during the tightening of monetary policy. To ascertain whether our findings are consistent across all periods, we divided the period of study as pre-crisis (2002 to 2007) and post-crisis period (2007-2015) and ran the same set of models. Interestingly, all liquidity variables are highly significant in the post-crisis period. However, the pre-crisis period has witnessed a moderate predictability of monetary policy. To check the robustness of our results we ran the same set of VAR models with different monetary policy variables and found the similar results. Unlike previous studies, we found most of the liquidity variables are significant throughout the sample period. This reveals the predictability of monetary policy on aggregate market liquidity. This study contributes to the existing body of literature by documenting a strong predictability of monetary policy on stock liquidity in an emerging economy with an order driven market making system like India. Most of the previous studies have been carried out in developing economies with quote driven or hybrid market making system and their results are ambiguous across different periods. From an eclectic sense, this study may be considered as a baseline study to further find out the macroeconomic determinants of liquidity of stocks at individual as well as aggregate level.

Keywords: market liquidity, monetary policy, order driven market, VAR, vector autoregressive model

Procedia PDF Downloads 361

Authors: Sarka Langer, Despoina Teli, Blanka Cabovska, Jan-Olof Dalenbäck, Lars Ekberg, Gabriel Bekö, Pawel Wargocki, Natalia Giraldo Vasquez

Abstract:

The aim of the study was to investigate the impact of various ventilation systems on indoor climate, air pollution, chemistry, and perception. Measurements of thermal environment and indoor air quality were performed in 45 primary school classrooms in Gothenburg, Sweden. The classrooms were grouped into three categories according to their ventilation system: category A) natural or exhaust ventilation or automated window opening; category B) balanced mechanical ventilation systems with constant air volume (CAV); and category C) balanced mechanical ventilation systems with variable air volume (VAV). A questionnaire survey about indoor air quality, perception of temperature, odour, noise and light, and sensation of well-being, alertness focus, etc., was distributed among the 10-12 years old children attending the classrooms. The results (medians) showed statistically significant differences between ventilation category A and categories B and C, but not between categories B and C in air change rates, median concentrations of carbon dioxide, individual volatile organic compounds formaldehyde and isoprene, in-door-to-outdoor ozone ratios and products of ozonolysis of squalene, a constituent of human skin oils, 6-methyl-5-hepten-2-one and decanal. Median ozone concentration, ozone loss -a difference between outdoor and indoor ozone concentrations- were different only between categories A and C. Median concentration of total VOCs and a perception index based on survey responses on perceptions and sensations indoors were not significantly different. In conclusion, ventilation systems have an impact on air change rates, indoor air quality, and chemistry, but the Swedish primary school children’s perception did not differ with the ventilation systems of the classrooms.

Keywords: indoor air pollutants, indoor climate, indoor chemistry, air change rate, perception

Procedia PDF Downloads 44

Authors: Benjamin Chijioke Agwah, Paulinus Chinaenye Eze

Abstract:

Antilock braking system (ABS) is one of the important contributions by the automobile industry, designed to ensure road safety in such way that vehicles are kept steerable and stable when during emergency braking. This paper presents a wheel slip-based intelligent controller with variable zero lag compensation for ABS. It is required to achieve a very fast perfect wheel slip tracking during hard braking condition and eliminate chattering with improved transient and steady state performance, while shortening the stopping distance using effective braking torque less than maximum allowable torque to bring a braking vehicle to a stop. The dynamic of a vehicle braking with a braking velocity of 30 ms⁻¹ on a straight line was determined and modelled in MATLAB/Simulink environment to represent a conventional ABS system without a controller. Simulation results indicated that system without a controller was not able to track desired wheel slip and the stopping distance was 135.2 m. Hence, an intelligent control based on fuzzy logic controller (FLC) was designed with a variable zero lag compensator (VZLC) added to enhance the performance of FLC control variable by eliminating steady state error, provide improve bandwidth to eliminate the effect of high frequency noise such as chattering during braking. The simulation results showed that FLC- VZLC provided fast tracking of desired wheel slip, eliminate chattering, and reduced stopping distance by 70.5% (39.92 m), 63.3% (49.59 m), 57.6% (57.35 m) and 50% (69.13 m) on dry, wet, cobblestone and snow road surface conditions respectively. Generally, the proposed system used effective braking torque that is less than the maximum allowable braking torque to achieve efficient wheel slip tracking and overall robust control performance on different road surfaces.

Keywords: ABS, fuzzy logic controller, variable zero lag compensator, wheel slip tracking

Procedia PDF Downloads 132

Authors: Mohd Zubairy Bin Shamsudin

Abstract:

Work-related Musculoskeletal Disorder (WMSDs) is one of the occupational health problems encountered by workers over the world. In Malaysia, there is increasing in trend over the years, particularly in the manufacturing sectors. Current method to observe workplace WMSDs is self-report questionnaire, observation and direct measurement. Observational method is most frequently used by the researcher and practitioner because of the simplified, quick and versatile when it applies to the worksite. However, there are some limitations identified e.g. some approach does not cover a wide spectrum of biomechanics activity and not sufficiently sensitive to assess the actual risks. This paper elucidates the development of Work Movement Task Analysis (WMTA), which is an observational tool for industrial practitioners’ especially untrained personnel to assess WMSDs risk factors and provide a basis for suitable intervention. First stage of the development protocol involved literature reviews, practitioner survey, tool validation and reliability. A total of six themes/comments were received in face validity stage. New revision of WMTA consisted of four sections of postural (neck, back, shoulder, arms, and legs) and associated risk factors; movement, load, coupling and basic environmental factors (lighting, noise, odorless, heat and slippery floor). For inter-rater reliability study shows substantial agreement among rater with K = 0.70. Meanwhile, WMTA validation shows significant association between WMTA score and self-reported pain or discomfort for the back, shoulder&arms and knee&legs with p<0.05. This tool is expected to provide new workplace ergonomic observational tool to assess WMSDs for the next stage of the case study.

Keywords: assessment, biomechanics, musculoskeletal disorders, observational tools

Procedia PDF Downloads 454