Search results for: random coefficients model

Authors: Nikolaos Papadimas, Joanna Bennett, Amir Sakhaei, Timothy Dodwell

Abstract:

Composite modeling of consolidation processes is playing an important role in the process and part design by indicating the formation of possible unwanted prior to expensive experimental iterative trial and development programs. Composite materials in their uncured state display complex constitutive behavior, which has received much academic interest, and this with different models proposed. Errors from modeling and statistical which arise from this fitting will propagate through any simulation in which the material model is used. A general hyperelastic polynomial representation was proposed, which can be readily implemented in various nonlinear finite element packages. In our case, FEniCS was chosen. The coefficients are assumed uncertain, and therefore the distribution of parameters learned using Markov Chain Monte Carlo (MCMC) methods. In engineering, the approach often followed is to select a single set of model parameters, which on average, best fits a set of experiments. There are good statistical reasons why this is not a rigorous approach to take. To overcome these challenges, A hierarchical Bayesian framework was proposed in which population distribution of model parameters is inferred from an ensemble of experiments tests. The resulting sampled distribution of hyperparameters is approximated using Maximum Entropy methods so that the distribution of samples can be readily sampled when embedded within a stochastic finite element simulation. The methodology is validated and demonstrated on a set of consolidation experiments of AS4/8852 with various stacking sequences. The resulting distributions are then applied to stochastic finite element simulations of the consolidation of curved parts, leading to a distribution of possible model outputs. With this, the paper, as far as the authors are aware, represents the first stochastic finite element implementation in composite process modelling.

Keywords: data-driven , material consolidation, stochastic finite elements, surrogate models

Procedia PDF Downloads 142

Authors: Boguslaw Schreyer

Abstract:

During the wheeled robot’s braking process, the extra dynamic vertical forces act on all wheels: left, right, front or rear. Those forces are directed downward on the front wheels while directed upward on the rear wheels. In order to maximize the deceleration, therefore, minimize the braking time and braking distance, we need to calculate a correct torque distribution: the front braking torque should be increased, and rear torque should be decreased. At the same time, we need to provide better transversal stability. In a simple case of all adhesion coefficients being the same under all wheels, the torque distribution may secure the optimal (maximal) control of the robot braking process, securing the minimum braking distance and a minimum braking time. At the same time, the transversal stability is relatively good. At any time, we control the transversal acceleration. In the case of the transversal movement, we stop the braking process and re-apply braking torque after a defined period of time. If we correctly calculate the value of the torques, we may secure the traction coefficient under the front and rear wheels close to its maximum. Also, in order to provide an optimum braking control, we need to calculate the timing of the braking torque application and the timing of its release. The braking torques should be released shortly after the wheels passed a maximum traction coefficient (while a wheels’ slip increases) and applied again after the wheels pass a maximum of traction coefficient (while the slip decreases). The correct braking torque distribution secures the front and rear wheels, passing this maximum at the same time. It guarantees an optimum deceleration control, therefore, minimum braking time. In order to calculate a correct torque distribution, a control unit should receive the input signals of a rear torque value (which changes independently), the robot’s deceleration, and values of the vertical front and rear forces. In order to calculate the timing of torque application and torque release, more signals are needed: speed of the robot: angular speed, and angular deceleration of the wheels. In case of different adhesion coefficients under the left and right wheels, but the same under each pair of wheels- the same under right wheels and the same under left wheels, the Select-Low (SL) and select high (SH) methods are applied. The SL method is suggested if transversal stability is more important than braking efficiency. Often in the case of the robot, more important is braking efficiency; therefore, the SH method is applied with some control of the transversal stability. In the case that all adhesion coefficients are different under all wheels, the front-rear torque distribution is maintained as in all previous cases. However, the timing of the braking torque application and release is controlled by the rear wheels’ lowest adhesion coefficient. The Lagrange equations have been used to describe robot dynamics. Matlab has been used in order to simulate the process of wheeled robot braking, and in conclusion, the braking methods have been selected.

Keywords: wheeled robots, braking, traction coefficient, asymmetric

Procedia PDF Downloads 160

Authors: Mirza Mohibulla Baig, Imil Hamda Imran, Tri Bagus Susilo, Sami El Ferik

Abstract:

The paper deals with system identification and control a nonlinear model of semi-autonomous underwater vehicle (UUV). The input-output data is first generated using the experimental values of the model parameters and then this data is used to compute the estimated parameter values. In this study, we use the semi-autonomous UUV LAURS model, which is developed by the Sensors and Actuators Laboratory in University of Sao Paolo. We applied three methods to identify the parameters: integral method, which is a classical least square method, recursive least square, and weighted recursive least square. In this paper, we also apply three different inputs (step input, sine wave input and random input) to each identification method. After the identification stage, we investigate the control performance of yaw motion of nonlinear semi-autonomous Unmanned Underwater Vehicle (UUV) using feedback linearization-based controller. In addition, we compare the performance of the control with an integral and a non-integral part along with state feedback. Finally, disturbance rejection and resilience of the controller is tested. The results demonstrate the ability of the system to recover from such fault.

Keywords: system identification, underwater vehicle, integral method, recursive least square, weighted recursive least square, feedback linearization, integral error

Procedia PDF Downloads 529

Authors: Kanchan Mondal, Dasharath Koulage, Dattatray Manerikar, Asmita Ghate

Abstract:

This paper presents an additive reliability model using warranty data and Finite Element Analysis (FEA) data. Warranty data for any product gives insight to its underlying issues. This is often used by Reliability Engineers to build prediction model to forecast failure rate of parts. But there is one major limitation in using warranty data for prediction. Warranty periods constitute only a small fraction of total lifetime of a product, most of the time it covers only the infant mortality and useful life zone of a bathtub curve. Predicting with warranty data alone in these cases is not generally provide results with desired accuracy. Failure rate of a mechanical part is driven by random issues initially and wear-out or usage related issues at later stages of the lifetime. For better predictability of failure rate, one need to explore the failure rate behavior at wear out zone of a bathtub curve. Due to cost and time constraints, it is not always possible to test samples till failure, but FEA-Fatigue analysis can provide the failure rate behavior of a part much beyond warranty period in a quicker time and at lesser cost. In this work, the authors proposed an Additive Weibull Model, which make use of both warranty and FEA fatigue analysis data for predicting failure rates. It involves modeling of two data sets of a part, one with existing warranty claims and other with fatigue life data. Hazard rate base Weibull estimation has been used for the modeling the warranty data whereas S-N curved based Weibull parameter estimation is used for FEA data. Two separate Weibull models’ parameters are estimated and combined to form the proposed Additive Weibull Model for prediction.

Keywords: bathtub curve, fatigue, FEA, reliability, warranty, Weibull

Procedia PDF Downloads 68

Authors: Marina Yurievna Aleksandrova

Abstract:

Research of the Holocaust retains value not only for history but also for sociology and psychology. One of the most important fields of study is how people were coping during and after this traumatic event. The aim of this paper is to identify the main contexts of the topic of love and to determine which contexts are more characteristic for different groups of victims of the Holocaust (gender, nationality, age). In this research, transcripts of interviews with Holocaust victims that were collected during 1946 for the "Voices of the Holocaust" project were used as data. Main contexts were analyzed with methods of network analysis and latent semantic analysis and classified by gender, age, and nationality with random forest. The results show that love is articulated and described significantly differently for male and female informants, nationality is shown results with lower values of quality metrics, as well as the age.

Keywords: Holocaust, latent semantic analysis, network analysis, text-mining, random forest

Procedia PDF Downloads 178

Authors: Medjadj Tarek, Ghribi Hayet

Abstract:

This study aims to develop a flood sensitivity assessment tool using machine learning (ML) techniques and geographic information system (GIS). The importance of this study is integrating the geographic information systems (GIS) and machine learning (ML) techniques for mapping flood risks, which help decision-makers to identify the most vulnerable areas and take the necessary precautions to face this type of natural disaster. To reach this goal, we will study the case of the city of M'sila, which is among the areas most vulnerable to floods. This study drew a map of flood-prone areas based on the methodology where we have made a comparison between 3 machine learning algorithms: the xGboost model, the Random Forest algorithm and the K Nearest Neighbour algorithm. Each of them gave an accuracy respectively of 97.92 - 95 - 93.75. In the process of mapping flood-prone areas, the first model was relied upon, which gave the greatest accuracy (xGboost).

Keywords: Geographic information systems (GIS), machine learning (ML), emergency mapping, flood disaster management

Procedia PDF Downloads 87

Authors: Go Bong Choi, Shin Je Lee, Sung Jin Yoo, Gibaek Lee, Jong Min Lee

Abstract:

In South Korea, it is difficult to obtain data for statistical pipe assessment. In this paper, to address these issues, we find that various statistical model presented before is how data mixed with noise and are whether apply in South Korea. Three major type of model is studied and if data is presented in the paper, we add noise to data, which affects how model response changes. Moreover, we generate data from model in paper and analyse effect of noise. From this we can find robustness and applicability in Korea of each model.

Keywords: proportional hazard model, survival model, water main deterioration, ecological sciences

Procedia PDF Downloads 739

Authors: Hadri Hussain, M.Nasir Ibrahim, Chee-Ming Ting, Mariani Idroas, Fuad Numan, Alias Mohd Noor

Abstract:

A biometric security system has become an important application in client identification and verification system. A conventional biometric system is normally based on unimodal biometric that depends on either behavioural or physiological information for authentication purposes. The behavioural biometric depends on human body biometric signal (such as speech) and biosignal biometric (such as electrocardiogram (ECG) and phonocardiogram or heart sound (HS)). The speech signal is commonly used in a recognition system in biometric, while the ECG and the HS have been used to identify a person’s diseases uniquely related to its cluster. However, the conventional biometric system is liable to spoof attack that will affect the performance of the system. Therefore, a multimodal biometric security system is developed, which is based on biometric signal of ECG, HS, and speech. The biosignal data involved in the biometric system is initially segmented, with each segment Mel Frequency Cepstral Coefficients (MFCC) method is exploited for extracting the feature. The Hidden Markov Model (HMM) is used to model the client and to classify the unknown input with respect to the modal. The recognition system involved training and testing session that is known as client identification (CID). In this project, twenty clients are tested with the developed system. The best overall performance at 44 kHz was 93.92% for ECG and the worst overall performance was ECG at 88.47%. The results were compared to the best overall performance at 44 kHz for (20clients) to increment of clients, which was 90.00% for HS and the worst overall performance falls at ECG at 79.91%. It can be concluded that the difference multimodal biometric has a substantial effect on performance of the biometric system and with the increment of data, even with higher frequency sampling, the performance still decreased slightly as predicted.

Keywords: electrocardiogram, phonocardiogram, hidden markov model, mel frequency cepstral coeffiecients, client identification

Procedia PDF Downloads 275

Authors: Αlexios Kotsakis, Stefanos Katsavounis, Dimitra Alexiou

Abstract:

Networks can be utilized to represent project planning problems, using nodes for activities and arcs to indicate precedence relationship between them. For fixed activity duration, a simple algorithm calculates the amount of time required to complete a project, followed by the activities that comprise the critical path. Program Evaluation and Review Technique (PERT) generalizes the above model by incorporating uncertainty, allowing activity durations to be random variables, producing nevertheless a relatively crude solution in planning problems. In this paper, based on the findings of the relevant literature, which strongly suggests that a Beta distribution can be employed to model earthmoving activities, we utilize Monte Carlo simulation, to estimate the project completion time distribution and measure the influence of skewness, an element inherent in activities of modern technical projects. We also extract the activity criticality index, with an ultimate goal to produce more accurate planning estimations.

Keywords: beta distribution, PERT, Monte Carlo simulation, skewness, project completion time distribution

Procedia PDF Downloads 145

Authors: Hee-Jae Suh, Kyung-Su Kim, Min-Ji Kim, Yeon-Seong Jeong, Ok-Hwan Lee, Jae-Wook Shin, Hyang-Sook Chun, Chan Lee

Abstract:

Canthaxanthin is keto-carotenoid produced from beta-carotene and it has been approved to be used in many countries as a food coloring agent. Canthaxanthin has been analyzed using High Performance Liquid Chromatography (HPLC) system with various ways of pretreatment methods. Four official methods for verification of canthaxanthin at FSA (UK), AOAC (US), EFSA (EU) and MHLW (Japan) were compared to improve its analytical and the pretreatment method. The Linearity, the limit of detection (LOD), the limit of quantification (LOQ), the accuracy, the precision and the recovery ratio were determined from each method with modification in pretreatment method. All HPLC methods exhibited correlation coefficients of calibration curves for canthaxanthin as 0.9999. The analysis methods from FSA, AOAC, and MLHW showed the LOD of 0.395 ppm, 0.105 ppm, and 0.084 ppm, and the LOQ of 1.196 ppm, 0.318 ppm, 0.254 ppm, respectively. Among tested methods, HPLC method of MHLW with modification in pretreatments was finally selected for the analysis of canthaxanthin in lab, because it exhibited the resolution factor of 4.0 and the selectivity of 1.30. This analysis method showed a correlation coefficients value of 0.9999 and the lowest LOD and LOQ. Furthermore, the precision ratio was lower than 1 and the accuracy was almost 100%. The method presented the recovery ratio of 90-110% with modification in pretreatment method. The cross-validation of coefficient variation was 5 or less among tested three institutions in Korea.

Keywords: analytic method, canthaxanthin, food colorants, pretreatment method

Procedia PDF Downloads 678

Authors: O. J. Akinsola, Yinan Zheng, Rose Anorlu, F. T. Ogunsola, Lifang Hou, Robert Leo-Murphy

Abstract:

Cervical Cancer (CC) is the 2nd most common cancer among women living in low and middle-income countries, with no associated symptoms during formative periods. With the advancement and innovative medical research, there are numerous preventive measures being utilized, but the incidence of cervical cancer cannot be truncated with the application of only screening tests. The mortality associated with this invasive cervical cancer can be nipped in the bud through the important role of early-stage detection. This study research selected an array of different top features selection techniques which was aimed at developing a model that could validly diagnose the risk factors of cervical cancer. A retrospective clinic-based cohort study was conducted on 178 HIV-associated cervical cancer patients in Lagos University teaching Hospital, Nigeria (U54 data repository) in April 2022. The outcome measure was the automated prediction of the HIV-associated cervical cancer cases, while the predictor variables include: demographic information, reproductive history, birth control, sexual history, cervical cancer screening history for invasive cervical cancer. The proposed technique was assessed with R and Python programming software to produce the model by utilizing the classification algorithms for the detection and diagnosis of cervical cancer disease. Four machine learning classification algorithms used are: the machine learning model was split into training and testing dataset into ratio 80:20. The numerical features were also standardized while hyperparameter tuning was carried out on the machine learning to train and test the data. Logistic Regression (LR), Decision Tree (DT), Random Forest (RF), and K-Nearest Neighbor (KNN). Some fitting features were selected for the detection and diagnosis of cervical cancer diseases from selected characteristics in the dataset using the contribution of various selection methods for the classification cervical cancer into healthy or diseased status. The mean age of patients was 49.7±12.1 years, mean age at pregnancy was 23.3±5.5 years, mean age at first sexual experience was 19.4±3.2 years, while the mean BMI was 27.1±5.6 kg/m2. A larger percentage of the patients are Married (62.9%), while most of them have at least two sexual partners (72.5%). Age of patients (OR=1.065, p<0.001**), marital status (OR=0.375, p=0.011**), number of pregnancy live-births (OR=1.317, p=0.007**), and use of birth control pills (OR=0.291, p=0.015**) were found to be significantly associated with HIV-associated cervical cancer. On top ten 10 features (variables) considered in the analysis, RF claims the overall model performance, which include: accuracy of (72.0%), the precision of (84.6%), a recall of (84.6%) and F1-score of (74.0%) while LR has: an accuracy of (74.0%), precision of (70.0%), recall of (70.0%) and F1-score of (70.0%). The RF model identified 10 features predictive of developing cervical cancer. The age of patients was considered as the most important risk factor, followed by the number of pregnancy livebirths, marital status, and use of birth control pills, The study shows that data mining techniques could be used to identify women living with HIV at high risk of developing cervical cancer in Nigeria and other sub-Saharan African countries.

Keywords: associated cervical cancer, data mining, random forest, logistic regression

Procedia PDF Downloads 82

Authors: Nigar Hussain

Abstract:

It is a subset of big data. Dark data means those data in which we fail to use for future decisions. There are many issues in existing work, but some need powerful tools for utilizing dark data. It needs sufficient techniques to deal with dark data. That enables users to exploit their excellence, adaptability, speed, less time utilization, execution, and accessibility. Another issue is the way to utilize dark data to extract helpful information to settle on better choices. In this paper, we proposed upgrade strategies to remove the dark side from dark data. Using a supervised model and machine learning techniques, we utilized dark data and achieved an F1 score of 89.48%.

Keywords: big data, dark data, machine learning, heatmap, random forest

Procedia PDF Downloads 15

Authors: Sylvain Amailland, Jean-Hugh Thomas, Charles Pézerat, Romuald Boucheron, Jean-Claude Pascal

Abstract:

The noise requirements for naval and research vessels have seen an increasing demand for quieter ships in order to fulfil current regulations and to reduce the effects on marine life. Hence, new methods dedicated to the characterization of propeller noise, which is the main source of noise in the far-field, are needed. The study of cavitating propellers in closed-section is interesting for analyzing hydrodynamic performance but could involve significant difficulties for hydroacoustic study, especially due to reverberation and boundary layer noise in the tunnel. The aim of this paper is to present a numerical methodology for the identification of hydroacoustic sources on marine propellers using hydrophone arrays in a large hydrodynamic tunnel. The main difficulties are linked to the reverberation of the tunnel and the boundary layer noise that strongly reduce the signal-to-noise ratio. In this paper it is proposed to estimate the reflection coefficients using an inverse method and some reference transfer functions measured in the tunnel. This approach allows to reduce the uncertainties of the propagation model used in the inverse problem. In order to reduce the boundary layer noise, a cleaning algorithm taking advantage of the low rank and sparse structure of the cross-spectrum matrices of the acoustic and the boundary layer noise is presented. This approach allows to recover the acoustic signal even well under the boundary layer noise. The improvement brought by this method is visible on acoustic maps resulting from beamforming and DAMAS algorithms.

Keywords: acoustic imaging, boundary layer noise denoising, inverse problems, model adaptation

Procedia PDF Downloads 329

Authors: Sobia Mushtaq, Firdaus E. Bareen, Asma Tayyeb

Abstract:

This study conducted to investigate the metal biosorption potential of indigenous Trichoderma species (T. harzianum KS05T01, T. longibrachiatum KS09T03, Trichoderma sp KS17T09., T. viridi KS17T011, T. atrobruneo KS21T014, and T. citrinoviride) that have been isolated from contaminated soil of Kasur Tannery Waste Management Agency. The effect of different biosorption parameters as initial metal ion concentration, pH, contact time , and temperature of incubation was investigated on the biosorption potential of these species. The metal removal efficiency and (E%) and metal uptake capacity (mg/g) increased along with the increase of initial metal concentration in media. The Trichoderma species can tolerate and survive under heavy metal stress up to 800mg/L. Among the two isotherm models were applied on the biosorption data, Langmuir isotherm model and Freundlich isotherm model, maximum correlation coefficients values (R 2 ) of 1was found for Langmuir model, which showed the better fitted model for the Trichoderma biosorption. The metal biosorption was increased with the increase of temperature and pH of the media. The maximum biosorption was observed between 25-30 o C and at pH 6.-7.5, while the biosorption rate was increased from 3-6 days of incubation, and then the rate of biosorption was slowed down. The biosorption data was better fitted for Pseudo kinetic first order during the initial days of biosorption. Thermodynamic parameters as standard Gibbs free energy (G), standard enthalpy change (H), and standard entropy (S) were calculated. The results confirmed the heavy metal biosorption by Trichoderma species was endothermic and spontaneous reaction in nature. The FTIR spectral analysis and SEM-EDX analysis of the treated and controlled mycelium revealed the changes in the active functional sites and morphological variations of the outer surface. The data analysis envisaged that high metal tolerance exhibited by Trichoderma species indicates its potential as efficacious and successful mediator for bioremediation of the heavy metal polluted environments.

Keywords: heavy metal, fungal biomass, biosorption, kinetics

Procedia PDF Downloads 116

Authors: Kailash C. Madan

Abstract:

We study the steady state behaviour of a batch arrival single server queue in which the first service with general service times is compulsory and the second service with general service times is optional. We term such a two phase service as generalized Coxian-2 service. Just after completion of a service the server must take a vacation of random length of time with general vacation times. We obtain steady state probability generating functions for the queue size as well as the steady state mean queue size at a random epoch of time in explicit and closed forms. Some particular cases of interest including some known results have been derived.

Keywords: batch arrivals, compound Poisson process, generalized Coxian-2 service, steady state

Procedia PDF Downloads 454

Authors: Bruna P. De Souza, Ricardo C. De Almeida

Abstract:

In some cities in the State of Parana – Brazil and in other countries atmospheric bromeliads (Tillandsia spp - Bromeliaceae) are considered weeds in trees, electric power lines, satellite dishes and other artificial supports. In this study, a numerical model was developed to simulate the seed dispersal of the Tillandsia recurvata species by wind with the objective of evaluating seeds displacement in the city of Ponta Grossa – PR, Brazil, since it is considered that the region is already infested. The model simulates the dispersal of each individual seed integrating parameters from the atmospheric boundary layer (ABL) and the local wind, simulated by the Weather Research Forecasting (WRF) mesoscale atmospheric model for the 2012 to 2015 period. The dispersal model also incorporates the approximate number of bromeliads and source height data collected from most infested electric power lines. The seeds terminal velocity, which is an important input data but was not available in the literature, was measured by an experiment with fifty-one seeds of Tillandsia recurvata. Wind is the main dispersal agent acting on plumed seeds whereas atmospheric turbulence is a determinant factor to transport the seeds to distances beyond 200 meters as well as to introduce random variability in the seed dispersal process. Such variability was added to the model through the application of an Inverse Fast Fourier Transform to wind velocity components energy spectra based on boundary-layer meteorology theory and estimated from micrometeorological parameters produced by the WRF model. Seasonal and annual wind means were obtained from the surface wind data simulated by WRF for Ponta Grossa. The mean wind direction is assumed to be the most probable direction of bromeliad seed trajectory. Moreover, the atmospheric turbulence effect and dispersal distances were analyzed in order to identify likely regions of infestation around Ponta Grossa urban area. It is important to mention that this model could be applied to any species and local as long as seed’s biological data and meteorological data for the region of interest are available.

Keywords: atmospheric turbulence, bromeliad, numerical model, seed dispersal, terminal velocity, wind

Procedia PDF Downloads 137

Authors: Yaping Zhao, Yimin Zhang

Abstract:

In the present paper, the stationary random vibration of a uniform cantilever beam is investigated. Two types of damping mechanism, i.e. the external and internal viscous dampings, are taken into account simultaneously. The excitation form is the support motion, and it is ideal white. Because two type of damping mechanism are considered concurrently, the product of the modal damping ratio and the natural frequency is not a constant anymore. As a result, the infinite definite integral encountered in the process of computing the mean square response is more complex than that in the existing literature. One signal progress of this work is to have calculated these definite integrals accurately. The precise solution of the mean square response is thus obtained in the infinite series form finally. Numerical examples are supplied and the numerical outcomes acquired confirm the validity of the theoretical analyses.

Keywords: random vibration, cantilever beam, mean square response, white noise

Procedia PDF Downloads 380

Authors: Terefe Alemu

Abstract:

This study is intended to examine Ethiopia’s trade flow determinants and trade potential with G20 countries whether it was overtraded or there is/are trade potential by using trade gravity model. The sources of panel data used were IMF, WDI, United Nations population division, The Heritage Foundation, Washington's No. 1 think tank online website database, online distance calculator, and others for the duration of 2010 to 2019 for 10 consecutive years. The empirical data analyzing tool used was Random effect model (REM), which is effective in estimation of time-invariant data. The empirical data analyzed using STATA software result indicates that Ethiopia has a trade potential with seven countries of G20, whereas Ethiopia overtrade with 12 countries and EU region. The Ethiopia’s and G20 countries/region bilateral trade flow statistically significant/ p<0.05/determinants were the population of G20 countries, growth domestic products of G20 countries, growth domestic products of Ethiopia, geographical distance between Ethiopia and G20 countries. The top five G20 countries exported to Ethiopia were china, United State of America, European Union, India, and South Africa, whereas the top five G20 countries imported from Ethiopia were EU, China, United State of America, Saudi Arabia, and Germany, respectively. Finally, the policy implication were Ethiopia has to Keep the consistence of trade flow with overtraded countries and improve with under traded countries through trade policy revision, and secondly, focusing on the trade determinants to improve trade flow is recommended.

Keywords: trade gravity model, trade determinants, G20, international trade, trade potential

Procedia PDF Downloads 208

Authors: Alanoud Moraya Aldalan, Abdulaziz Almaleh

Abstract:

A crucial component of maintaining a customer-oriented business as in the telecom industry is understanding the reasons and factors that lead to customer churn. Competition between telecom companies has greatly increased in recent years. It has become more important to understand customers’ needs in this strong market of telecom industries, especially for those who are looking to turn over their service providers. So, predictive churn is now a mandatory requirement for retaining those customers. Machine learning can be utilized to accomplish this. Churn Prediction has become a very important topic in terms of machine learning classification in the telecommunications industry. Understanding the factors of customer churn and how they behave is very important to building an effective churn prediction model. This paper aims to predict churn and identify factors of customers’ churn based on their past service usage history. Aiming at this objective, the study makes use of feature selection, normalization, and feature engineering. Then, this study compared the performance of four different machine learning algorithms on the Orange dataset: Logistic Regression, Random Forest, Decision Tree, and Gradient Boosting. Evaluation of the performance was conducted by using the F1 score and ROC-AUC. Comparing the results of this study with existing models has proven to produce better results. The results showed the Gradients Boosting with feature selection technique outperformed in this study by achieving a 99% F1-score and 99% AUC, and all other experiments achieved good results as well.

Keywords: machine learning, gradient boosting, logistic regression, churn, random forest, decision tree, ROC, AUC, F1-score

Procedia PDF Downloads 131

Authors: Sajjad Asefi, Hossein Afrakhte

Abstract:

This paper presents the evaluation of reliability indices of an electrical distribution system using Monte Carlo simulation technique accounting Time To Switch (TTS) for each section. In this paper, the distribution system has been assumed by accounting random repair time omission. For simplicity, we have assumed the reliability analysis to be based on exponential law. Each segment has a specified rate of failure (λ) and repair time (r) which will give us the mean up time and mean down time of each section in distribution system. After calculating the modified mean up time (MUT) in years, mean down time (MDT) in hours and unavailability (U) in h/year, TTS have been added to the time which the system is not available, i.e. MDT. In this paper, we have assumed the TTS to be a random variable with Log-Normal distribution.

Keywords: distribution system, Monte Carlo simulation, reliability, repair time, time to switch (TTS)

Procedia PDF Downloads 421

Authors: Amit Gill, Lalith Wijerathne, Sebastian Poledna

Abstract:

This research presents the details of the implementation of high performance computing (HPC) extension of agent-based economic models (ABEMs) to simulate hundreds of millions of heterogeneous agents. ABEMs offer an alternative approach to study the economy as a dynamic system of interacting heterogeneous agents, and are gaining popularity as an alternative to standard economic models. Over the last decade, ABEMs have been increasingly applied to study various problems related to monetary policy, bank regulations, etc. When it comes to predicting the effects of local economic disruptions, like major disasters, changes in policies, exogenous shocks, etc., on the economy of the country or the region, it is pertinent to study how the disruptions cascade through every single economic entity affecting its decisions and interactions, and eventually affect the economic macro parameters. However, such simulations with hundreds of millions of agents are hindered by the lack of HPC enhanced ABEMs. In order to address this, a scalable Distributed Memory Parallel (DMP) implementation of ABEMs has been developed using message passing interface (MPI). A balanced distribution of computational load among MPI-processes (i.e. CPU cores) of computer clusters while taking all the interactions among agents into account is a major challenge for scalable DMP implementations. Economic agents interact on several random graphs, some of which are centralized (e.g. credit networks, etc.) whereas others are dense with random links (e.g. consumption markets, etc.). The agents are partitioned into mutually-exclusive subsets based on a representative employer-employee interaction graph, while the remaining graphs are made available at a minimum communication cost. To minimize the number of communications among MPI processes, real-life solutions like the introduction of recruitment agencies, sales outlets, local banks, and local branches of government in each MPI-process, are adopted. Efficient communication among MPI-processes is achieved by combining MPI derived data types with the new features of the latest MPI functions. Most of the communications are overlapped with computations, thereby significantly reducing the communication overhead. The current implementation is capable of simulating a small open economy. As an example, a single time step of a 1:1 scale model of Austria (i.e. about 9 million inhabitants and 600,000 businesses) can be simulated in 15 seconds. The implementation is further being enhanced to simulate 1:1 model of Euro-zone (i.e. 322 million agents).

Keywords: agent-based economic model, high performance computing, MPI-communication, MPI-process

Procedia PDF Downloads 123

Authors: Lou-Hon Sun, Yu-Ming Chiu, Chen-Wei Tao, Chia-Yun Tsai

Abstract:

It is common practice for many companies to ask employees to provide heart-touching service for customers and to emphasize the attitude of 'customer first'. However, services may not necessarily gain praise, and may actually be considered excessive, if customers do not appreciate such behaviors. In reality, many restaurant businesses try to provide as much service as possible without taking into account whether over-provision may lead to negative customer reception. A survey of 894 people in Britain revealed that 49 percent of respondents consider over-attentive waiters the most annoying aspect of dining out. It can be seen that merely aiming to exceed customers’ expectations without actually addressing their needs, only further distances and dissociates the standard of services from the goals of customer satisfaction itself. Over-service is defined, as 'service provided that exceeds customer expectations, or simply that customers deemed redundant, resulting in negative perception'. It was found that customers’ reactions and complaints concerning over-service are not as intense as those against service failures caused by the inability to meet expectations; consequently, it is more difficult for managers to become aware of the existence of over-service. Thus the ability to manage over-service behaviors is a significant topic for consideration. The Kano model classifies customer preferences into five categories: attractive quality attribute, one-dimensional quality attribute, must-be quality attribute, indifferent quality attribute and reverse quality attributes. The model is still very popular for researchers to explore the quality aspects and customer satisfaction. Nevertheless, several studies indicated that Kano’s model could not fully capture the nature of service quality. The concept of over-service can be used to restructure the model and provide a better understanding of the service quality construct. In this research, the structure of Kano's two-dimensional questionnaire will be used to classify the factors into different dimensions. The same questions will be used in the second questionnaire for identifying the over-service experienced of the respondents. The finding of these two questionnaires will be used to analyze the relevance between service quality classification and over-service behaviors. The subjects of this research are customers of fine dining chain restaurants. Three hundred questionnaires will be issued based on the stratified random sampling method. Items for measurement will be derived from DINESERV scale. The tangible dimension of the questionnaire will be eliminated due to this research is focused on the employee behaviors. Quality attributes of the Kano model are often regarded as an instrument for improving customer satisfaction. The concept of over-service can be used to restructure the model and provide a better understanding of service quality construct. The extension of the Kano model will not only develop a better understanding of customer needs and expectations but also enhance the management of service quality.

Keywords: consumer satisfaction, DINESERV, kano model, over-service

Procedia PDF Downloads 159

Authors: Thomas Chinwe Urama, Patrick Oseloka Ezepue, Peters Chimezie Nnanwa

Abstract:

This paper investigates the universal financial dynamics in two dominant stock markets in Sub-Saharan Africa, through an in-depth analysis of the cross-correlation matrix of price returns in Nigerian Stock Market (NSM) and Johannesburg Stock Exchange (JSE), for the period 2009 to 2013. The strength of correlations between stocks is known to be higher in JSE than that of the NSM. Particularly important for modelling Nigerian derivatives in the future, the interactions of other stocks with the oil sector are weak, whereas the banking sector has strong positive interactions with the other sectors in the stock exchange. For the JSE, it is the oil sector and beverages that have greater sectorial correlations, instead of the banks which have the weaker correlation with other sectors in the stock exchange.

Keywords: random matrix theory, cross-correlations, emerging markets, option pricing, eigenvalues eigenvectors, inverse participation ratios and implied volatility

Procedia PDF Downloads 294

Authors: Jerome Joshi

Abstract:

The paper presents the modelling of financial markets using the Stochastic Pi Calculus model. The Stochastic Pi Calculus model is mainly used for biological applications; however, the feature of this model promotes its use in financial markets, more prominently in high frequency trading. The trading system can be broadly classified into exchange, market makers or intermediary traders and fundamental traders. The exchange is where the action of the trade is executed, and the two types of traders act as market participants in the exchange. High frequency trading, with its complex networks and numerous market participants (intermediary and fundamental traders) poses a difficulty while modelling. It involves the participants to seek the advantage of complex trading algorithms and high execution speeds to carry out large volumes of trades. To earn profits from each trade, the trader must be at the top of the order book quite frequently by executing or processing multiple trades simultaneously. This would require highly automated systems as well as the right sentiment to outperform other traders. However, always being at the top of the book is also not best for the trader, since it was the reason for the outbreak of the ‘Hot – Potato Effect,’ which in turn demands for a better and more efficient model. The characteristics of the model should be such that it should be flexible and have diverse applications. Therefore, a model which has its application in a similar field characterized by such difficulty should be chosen. It should also be flexible in its simulation so that it can be further extended and adapted for future research as well as be equipped with certain tools so that it can be perfectly used in the field of finance. In this case, the Stochastic Pi Calculus model seems to be an ideal fit for financial applications, owing to its expertise in the field of biology. It is an extension of the original Pi Calculus model and acts as a solution and an alternative to the previously flawed algorithm, provided the application of this model is further extended. This model would focus on solving the problem which led to the ‘Flash Crash’ which is the ‘Hot –Potato Effect.’ The model consists of small sub-systems, which can be integrated to form a large system. It is designed in way such that the behavior of ‘noise traders’ is considered as a random process or noise in the system. While modelling, to get a better understanding of the problem, a broader picture is taken into consideration with the trader, the system, and the market participants. The paper goes on to explain trading in exchanges, types of traders, high frequency trading, ‘Flash Crash,’ ‘Hot-Potato Effect,’ evaluation of orders and time delay in further detail. For the future, there is a need to focus on the calibration of the module so that they would interact perfectly with other modules. This model, with its application extended, would provide a basis for researchers for further research in the field of finance and computing.

Keywords: concurrent computing, high frequency trading, financial markets, stochastic pi calculus

Procedia PDF Downloads 76

Authors: Saheed Olakunle Sanusi, Israel Ajibade Adedeji

Abstract:

The study analysed the effect of microfinance on the profit level of small and medium scale enterprises in Lagos. The data for the study were obtained by simple random sampling, and total of one hundred and fifty (150) small and medium scale enterprises (SMEs) were sampled for the study. Seventy-five (75) each are microfinance users and non-users. Data were analysed using descriptive statistics, logit model, t-test and ordinary least square (OLS) regression. The mean profit of the enterprises using microfinance is ₦16.8m, while for the non-users of microfinance is ₦5.9m. The mean profit of microfinance users is statistically different from the non-users. The result of the logit model specified for the determinant of access to microfinance showed that three of specified variables- educational status of the enterprise head, credit utilisation and volume of business investment are significant at P < 0.01. Enterprises with many years of experience, highly educated enterprise heads and high volume of business investment have more potential access to microfinance. The OLS regression model indicated that three parameters namely number of school years, the volume of business investment and (dummy) participation in microfinance were found to be significant at P < 0.05. These variables are therefore significant determinants of impacts of microfinance on profit level in the study area. The study, therefore, concludes and recommends that to improve the status of small and medium scale enterprises for an increase in profit, the full benefit of access to microfinance can be enhanced through investment in social infrastructure and human capital development. Also, concerted efforts should be made to encouraged non-users of microfinance among SMEs to use it in order to boost their profit.

Keywords: credit utilisation, logit model, microfinance, small and medium enterprises

Procedia PDF Downloads 200

Authors: Xiao Chen, Xiaoying Kong, Min Xu

Abstract:

This paper presents a road vehicle detection approach for the intelligent transportation system. This approach mainly uses low-cost magnetic sensor and associated data collection system to collect magnetic signals. This system can measure the magnetic field changing, and it also can detect and count vehicles. We extend Mel Frequency Cepstral Coefficients to analyze vehicle magnetic signals. Vehicle type features are extracted using representation of cepstrum, frame energy, and gap cepstrum of magnetic signals. We design a 2-dimensional map algorithm using Vector Quantization to classify vehicle magnetic features to four typical types of vehicles in Australian suburbs: sedan, VAN, truck, and bus. Experiments results show that our approach achieves a high level of accuracy for vehicle detection and classification.

Keywords: vehicle classification, signal processing, road traffic model, magnetic sensing

Procedia PDF Downloads 314

Authors: H. Bensouilah, H. Boucherit, M. Lahmar

Abstract:

A theoretical investigation on the effects of both steady-state and dynamic deformations of the foils on the dynamic performance characteristics of a self-acting air foil journal bearing operating under small harmonic vibrations is proposed. To take into account the dynamic deformations of foils, the perturbation method is used for determining the gas-film stiffness and damping coefficients for given values of excitation frequency, compressibility number, and compliance factor of the bump foil. The nonlinear stationary Reynolds’ equation is solved by means of the Galerkins’ finite element formulation while the finite differences method are used to solve the first order complex dynamic equations resulting from the perturbation of the nonlinear transient compressible Reynolds’ equation. The stiffness of a bump is uniformly distributed throughout the bearing surface (generation I bearing). It was found that the dynamic properties of the compliant finite length journal bearing are significantly affected by the compliance of foils especially when the dynamic deformation of foils is considered in addition to the static one by applying the principle of superposition.

Keywords: elasto-aerodynamic lubrication, air foil bearing, steady-state deformation, dynamic deformation, stiffness and damping coefficients, perturbation method, fluid-structure interaction, Galerk infinite element method, finite difference method

Procedia PDF Downloads 390

Authors: Tunjo Perič, Marin Fatović

Abstract:

In this paper a new methodology for vendor selection and supply quotas determination (VSSQD) is proposed. The problem of VSSQD is solved by the model that combines revised weighting method for determining the objective function coefficients, and a multiple objective linear programming (MOLP) method based on the cooperative game theory for VSSQD. The criteria used for VSSQD are: (1) purchase costs and (2) product quality supplied by individual vendors. The proposed methodology is tested on the example of flour purchase for a bakery with two decision makers.

Keywords: cooperative game theory, multiple objective linear programming, revised weighting method, vendor selection

Procedia PDF Downloads 355

Authors: CarlyWill Sloan

Abstract:

Racial disparities in criminal justice outcomes are well-documented. However, there is little evidence on the extent to which racial bias by prosecutors is responsible for these disparities. This paper tests for racial bias in conviction by prosecutors. To identify effects, this paper leverages as good as random variation in prosecutor race using detailed administrative data on the case assignment process and case outcomes in New York County, New York. This paper shows that the assignment of an opposite-race prosecutor leads to a 5 percentage point (~ 8 percent) increase in the likelihood of conviction for property crimes. There is no evidence of effects for other types of crimes. Additional results indicate decreased dismissals by opposite-race prosecutors likely drive my property crime estimates.

Keywords: criminal justice, discrimination, prosecutors, racial disparities

Procedia PDF Downloads 189

Authors: Desplanches Maxime

Abstract:

Accurate estimation of the remaining useful life of lithium-ion batteries for electronic devices is crucial. Data-driven methodologies encounter challenges related to data volume and acquisition protocols, particularly in capturing a comprehensive range of aging indicators. To address these limitations, we propose a hybrid approach that integrates an electrochemical model with state-of-the-art data analysis techniques, yielding a comprehensive database. Our methodology involves infusing an aging phenomenon into a Newman model, leading to the creation of an extensive database capturing various aging states based on non-destructive parameters. This database serves as a robust foundation for subsequent analysis. Leveraging advanced data analysis techniques, notably principal component analysis and t-Distributed Stochastic Neighbor Embedding, we extract pivotal information from the data. This information is harnessed to construct a regression function using either random forest or support vector machine algorithms. The resulting predictor demonstrates a 5% error margin in estimating remaining battery life, providing actionable insights for optimizing usage. Furthermore, the database was built from the Newman model calibrated for aging and performance using data from a European project called Teesmat. The model was then initialized numerous times with different aging values, for instance, with varying thicknesses of SEI (Solid Electrolyte Interphase). This comprehensive approach ensures a thorough exploration of battery aging dynamics, enhancing the accuracy and reliability of our predictive model. Of particular importance is our reliance on the database generated through the integration of the electrochemical model. This database serves as a crucial asset in advancing our understanding of aging states. Beyond its capability for precise remaining life predictions, this database-driven approach offers valuable insights for optimizing battery usage and adapting the predictor to various scenarios. This underscores the practical significance of our method in facilitating better decision-making regarding lithium-ion battery management.

Keywords: Li-ion battery, aging, diagnostics, data analysis, prediction, machine learning, electrochemical model, regression

Procedia PDF Downloads 65