Search results for: multi criteria inventory classification models

Authors: Nima Pirhadi, Yong Bo Shao, Xusheng Wa, Jianguo Lu

Abstract:

Several models based on the simplified method introduced by Seed and Idriss (1971) have been developed to assess the liquefaction potential of saturated sandy soils. The procedure includes determining the cyclic resistance of the soil as the cyclic resistance ratio (CRR) and comparing it with earthquake loads as cyclic stress ratio (CSR). Of all methods to determine CRR, the methods using shear-wave velocity (Vs) are common because of their low sensitivity to the penetration resistance reduction caused by fine content (FC). To evaluate the capability of the models, based on the Vs., the new data from Bachu-Jianshi earthquake case history collected, then the prediction results of the models are compared to the measured results; consequently, the accuracy of the models are discussed via three criteria and graphs. The evaluation demonstrates reasonable accuracy of the models in the Banchu region.

Keywords: seismic liquefaction, banchu-jiashi earthquake, shear-wave velocity, liquefaction potential evaluation

Procedia PDF Downloads 205

Authors: Irina Khutsishvili

Abstract:

The presented study develops a decision support methodology for multi-criteria group decision-making problem. The proposed methodology is based on the TOPSIS (Technique for Order Performance by Similarity to Ideal Solution) approach in the hesitant fuzzy environment. The main idea of decision-making problem is a selection of one best alternative or several ranking alternatives among a set of feasible alternatives. Typically, the process of decision-making is based on an evaluation of certain criteria. In many MCDM problems (such as medical diagnosis, project management, business and financial management, etc.), the process of decision-making involves experts' assessments. These assessments frequently are expressed in fuzzy numbers, confidence intervals, intuitionistic fuzzy values, hesitant fuzzy elements and so on. However, a more realistic approach is using linguistic expert assessments (linguistic variables). In the proposed methodology both the values and weights of the criteria take the form of linguistic variables, given by all decision makers. Then, these assessments are expressed in triangular fuzzy numbers. Consequently, proposed approach is based on triangular hesitant fuzzy TOPSIS decision-making model. Following the TOPSIS algorithm, first, the fuzzy positive ideal solution (FPIS) and the fuzzy negative-ideal solution (FNIS) are defined. Then the ranking of alternatives is performed in accordance with the proximity of their distances to the both FPIS and FNIS. Based on proposed approach the software package has been developed, which was used to rank investment projects in the real investment decision-making problem. The application and testing of the software were carried out based on the data provided by the ‘Bank of Georgia’.

Keywords: fuzzy TOPSIS approach, investment project, linguistic variable, multi-criteria decision making, triangular hesitant fuzzy set

Procedia PDF Downloads 391

Authors: Jamal A. Radaideh, Ziad N. Shatnawi

Abstract:

Both indoor and outdoor air quality is investigated throughout residential areas of Al Hofuf city/ Eastern province of Saudi Arabia through a multi‐week multiple sites measurement and sampling survey. Concentration levels of five criteria air pollutants, including carbon dioxide (CO2), carbon monoxide (CO), nitrous dioxide (NO2), sulfur dioxide (SO2) and total volatile organic compounds (TVOC) were measured and analyzed during the study period from January to May 2014. For this survey paper, three different sites, roadside RS, urban UR, and rural RU were selected. Within each site type, six locations were assigned to carryout air quality measurements and to study varying indoor/outdoor air quality for each pollutant. Results indicate that a strong correlation between indoor and outdoor air exists. The I/O ratios for the considered criteria pollutants show that the strongest relationship between indoor and outdoor air is found by analyzing of carbon dioxide, CO2 (0.88), while the lowest is found by both NO2 and SO2 (0.7).

Keywords: criteria air pollutants, indoor/outdoor air pollution, indoor/outdoor ratio, Saudi Arabia

Procedia PDF Downloads 394

Authors: Arezoo Heidary, Abolfazl Mirzazadeh, Aref Gholami-Qadikolaei

Abstract:

In the business environment it is very common that a lot received may not be equal to quantity ordered. in this work, a random disturbance in a received quantity is considered. It is assumed a maximum allowable limit for storage space and inventory investment.The impact of lead time and ordering cost reductions once they act dependently is also investigated. Further, considering a mixture of back order and lost sales for allowable shortage system, the effect of investment on reducing lost sale rate is analyzed. For the proposed control system, a Lagrangian method is applied in order to solve the problem and an algorithmic procedure is utilized to achieve optimal solution with the global minimum expected cost. Finally, proves on concavity and convexity of the model in the decision variables are shown.

Keywords: stochastic inventory system, lead time, ordering cost, lost sale rate, inventory constraints, random disturbance

Procedia PDF Downloads 390

Authors: Galal Elkobrosy, Amr M. Abdelrazek, Bassuny M. Elsouhily, Mohamed E. Khidr

Abstract:

Crank shaft length, connecting rod length, crank angle, engine rpm, cylinder bore, mass of piston and compression ratio are the inputs that can control the performance of the slider crank mechanism and then its efficiency. Several combinations of these seven inputs are used and compared. The throughput engine torque predicted by the simulation is analyzed through two different regression models, with and without interaction terms, developed according to multi-linear regression using LU decomposition to solve system of algebraic equations. These models are validated. A regression model in seven inputs including their interaction terms lowered the polynomial degree from 3^rd degree to 1^st degree and suggested valid predictions and stable explanations.

Keywords: design of experiments, regression analysis, SI engine, statistical modeling

Procedia PDF Downloads 159

Authors: Claire Limoge Schraen, Philippe Gueguen, Cedric Giry, Cedric Desprez, Frédéric Ragueneau

Abstract:

Mediterranean area is characterized by numerous monumental or vernacular masonry structures illustrating old ways of build and live. Those precious buildings are often poorly documented, present complex shapes and loadings, and are protected by the States, leading to legal constraints. This area also presents a moderate to high seismic activity. Even moderate earthquakes can be magnified by local site effects and cause collapse or significant damage. Moreover the structural resistance of masonry buildings, especially when less famous or located in rural zones has been generally lowered by many factors: poor maintenance, unsuitable restoration, ambient pollution, previous earthquakes. Recent earthquakes prove that any damage to these architectural witnesses to our past is irreversible, leading to the necessity of acting preventively. This means providing preventive assessments for hundreds of structures with no or few documents. In this context we want to propose a general method, based on hierarchized numerical models, to provide preliminary structural diagnoses at a regional scale, indicating whether more precise investigations and models are necessary for each building. To this aim, we adapt different tools, being developed such as photogrammetry or to be created such as a preprocessor starting from pictures to build meshes for a FEM software, in order to allow dynamic studies of the buildings of the panel. We made an inventory of 198 baroque chapels and churches situated in the French Alps. Then their structural characteristics have been determined thanks field surveys and the MicMac photogrammetric software. Using structural criteria, we determined eight types of churches and seven types for chapels. We studied their dynamical behavior thanks to CAST3M, using EC8 spectrum and accelerogramms of the studied zone. This allowed us quantifying the effect of the needed simplifications in the most sensitive zones and choosing the most effective ones. We also proposed threshold criteria based on the observed damages visible in the in situ surveys, old pictures and Italian code. They are relevant in linear models. To validate the structural types, we made a vibratory measures campaign using vibratory ambient noise and velocimeters. It also allowed us validating this method on old masonry and identifying the modal characteristics of 20 churches. Then we proceeded to a dynamic identification between numerical and experimental modes. So we updated the linear models thanks to material and geometrical parameters, often unknown because of the complexity of the structures and materials. The numerically optimized values have been verified thanks to the measures we made on the masonry components in situ and in laboratory. We are now working on non-linear models redistributing the strains. So we validate the damage threshold criteria which we use to compute the vulnerability curves of each defined structural type. Our actual results show a good correlation between experimental and numerical data, validating the final modeling simplifications and the global method. We now plan to use non-linear analysis in the critical zones in order to test reinforcement solutions.

Keywords: heritage structures, masonry numerical modeling, seismic vulnerability assessment, vibratory measure

Procedia PDF Downloads 469

Authors: Manaranjan Pradhan, Shailaja Grover, U. Dinesh Kumar

Abstract:

Farming community in India, as well as other parts of the world, is one of the highly stressed communities due to reasons such as increasing input costs (cost of seeds, fertilizers, pesticide), droughts, reduced revenue leading to farmer suicides. Lack of integrated farm advisory system in India adds to the farmers problems. Farmers need right information during the early stages of crop’s lifecycle to prevent damage and loss in revenue. In this paper, we use deep learning techniques to develop an early warning system for detection of crop diseases using images taken by farmers using their smart phone. The research work leads to building a smart assistant using analytics and big data which could help the farmers with early diagnosis of the crop diseases and corrective actions. The classical approach for crop disease management has been to identify diseases at crop level. Recently, ImageNet Classification using the convolutional neural network (CNN) has been successfully used to identify diseases at individual plant level. Our model uses convolution filters, max pooling, dense layers and dropouts (to avoid overfitting). The models are built for binary classification (healthy or not healthy) and multi class classification (identifying which disease). Transfer learning is used to modify the weights of parameters learnt through ImageNet dataset and apply them on crop diseases, which reduces number of epochs to learn. One shot learning is used to learn from very few images, while data augmentation techniques are used to improve accuracy with images taken from farms by using techniques such as rotation, zoom, shift and blurred images. Models built using combination of these techniques are more robust for deploying in the real world. Our model is validated using tomato crop. In India, tomato is affected by 10 different diseases. Our model achieves an accuracy of more than 95% in correctly classifying the diseases. The main contribution of our research is to create a personal assistant for farmers for managing plant disease, although the model was validated using tomato crop, it can be easily extended to other crops. The advancement of technology in computing and availability of large data has made possible the success of deep learning applications in computer vision, natural language processing, image recognition, etc. With these robust models and huge smartphone penetration, feasibility of implementation of these models is high resulting in timely advise to the farmers and thus increasing the farmers' income and reducing the input costs.

Keywords: analytics in agriculture, CNN, crop disease detection, data augmentation, image recognition, one shot learning, transfer learning

Procedia PDF Downloads 97

Authors: Jeaha Yang, Ahmed Khan, Donna L. Rodela, Mohammed A. Qaudeer

Abstract:

Traditional supply chain silos still exist today due to the difficulty of establishing trust between various partners and technological barriers across industries. Companies lose opportunities and revenue and inadvertently make poor business decisions resulting in further challenges. Blockchain technology can bring a new level of transparency through sharing information with a distributed ledger in a decentralized manner that creates a basis of trust for business. Blockchain is a loosely coupled, hub-style communication network in which trading partners can work indirectly with each other for simpler integration, but they work together through the orchestration of their supply chain operations under a coherent process that is developed jointly. A Blockchain increases efficiencies, lowers costs, and improves interoperability to strengthen and automate the supply chain management process while all partners share the risk. Blockchain ledger is built to track inventory lifecycle for supply chain transparency and keeps a journal of inventory movement for real-time reconciliation. State design patterns are used to capture the life cycle (behavior) of inventory management as a state machine for a common, transparent and coherent process which creates an opportunity for trading partners to become more responsive in terms of changes or improvements in process, reconcile discrepancies, and comply with internal governance and external regulations. It enables end-to-end, inter-company visibility at the unit level for more accurate demand planning with better insight into order fulfillment and replenishment.

Keywords: supply chain management, inventory trace-ability, perpetual inventory system, inventory lifecycle, blockchain, inventory consignment, supply chain transparency, digital thread, demand planning, hyper ledger fabric

Procedia PDF Downloads 64

Authors: Po-Jen Chen, Jian-Jiun Ding, Hung-Wei Hsu, Chien-Yao Wang, Jia-Ching Wang

Abstract:

A more accurate scene classification algorithm using label relations and the hierarchy neural network was developed in this work. In many classification algorithms, it is assumed that the labels are mutually exclusive. This assumption is true in some specific problems, however, for scene classification, the assumption is not reasonable. Because there are a variety of objects with a photo image, it is more practical to assign multiple labels for an image. In this paper, two label relations, which are exclusive relation and hierarchical relation, were adopted in the classification process to achieve more accurate multiple label classification results. Moreover, the hierarchy neural network (hierarchy NN) is applied to classify the image and the directed acyclic graph structure is used for predicting a more reasonable result which obey exclusive and hierarchical relations. Simulations show that, with these techniques, a much more accurate scene classification result can be achieved.

Keywords: convolutional neural network, label relation, hierarchy neural network, scene classification

Procedia PDF Downloads 428

Authors: Sarah Beaver, Renee Bryce

Abstract:

Audiograms detect hearing impairment, but missing values pose problems. This work explores imputations in an attempt to improve accuracy. This work implements Linear Regression, Lasso, Linear Support Vector Regression, Bayesian Ridge, K Nearest Neighbors (KNN), and Random Forest machine learning techniques to impute audiogram frequencies ranging from 125Hz to 8000Hz. The data contains patients who had or were candidates for cochlear implants. Accuracy is compared across two different Nested Cross-Validation k values. Over 4000 audiograms were used from 800 unique patients. Additionally, training on data combines and compares left and right ear audiograms versus single ear side audiograms. The accuracy achieved using Root Mean Square Error (RMSE) values for the best models for Random Forest ranges from 4.74 to 6.37. The R\textsuperscript{2} values for the best models for Random Forest ranges from .91 to .96. The accuracy achieved using RMSE values for the best models for KNN ranges from 5.00 to 7.72. The R\textsuperscript{2} values for the best models for KNN ranges from .89 to .95. The best imputation models received R\textsuperscript{2} between .89 to .96 and RMSE values less than 8dB. We also show that the accuracy of classification predictive models performed better with our best imputation models versus constant imputations by a two percent increase.

Keywords: machine learning, audiograms, data imputations, single imputations

Procedia PDF Downloads 56

Authors: Isil Coklar-Okutkan, Miray Akyunus

Abstract:

Victim sensitivity is a form of justice sensitivity that reflects the tendency to perceive injustice to one’s disadvantage. Victim sensitivity is considered as a dysfunctional trait that predicts anger, aggression, uncooperative behavior, depression and anxiety. Indeed, exploring the mechanism of association between victim sensitivity and anger is clinically important since it can lead to externalizing and internalizing problems. This study aims to investigate the moderating role of pathological narcissism in the relationship between victim sensitivity and anger rumination. Through testing different models where subtypes of narcissism and anger rumination components are included independently, the specific mechanism of different ruminative processes in anger is investigated. The sample consisted of 311 undergraduate students from Turkey, 107 of whom were males, and 204 were females. Participants completed Justice Sensitivity Inventory-Victim Subscale, Pathological Narcissism Inventory and Anger Rumination Scale. In the proposed double moderation model, vulnerable and grandiose narcissism was the moderators in the relationship between victim justice sensitivity and anger rumination. Four separate models were tested where one of the four components of anger rumination (angry afterthoughts, thoughts of revenge, angry memories, understanding of causes) were the dependent variable in each model. Results revealed that two of the moderation models are significant. Firstly, grandiose narcissism is the only moderator in the relationship between victim sensitivity and thoughts of revenge. Secondly, vulnerable narcissism is the only moderator in the relationship between victim sensitivity and understanding causes. Accordingly, grandiose narcissism is positively associated with the thoughts of revenge, and vulnerable narcissism is positively associated with understanding causes, only when the level of victim sensitivity is high. To summarize, increased victim sensitivity leads to ruminative thoughts of revenge in individuals with grandiose narcissism, whereas it leads to rumination on causes of the incident in individuals with vulnerable narcissism. The clinical implications of the findings are discussed.

Keywords: anger rumination, victim sensitivity, grandiose narcissism, vulnerable narcissism

Procedia PDF Downloads 165

Authors: Fatima Z. Cherhabil, Mammar Sedrati, Sonia-Sabrina Bendib‎

Abstract:

In order to support the continued growth, critical latency of ‎IoT ‎applications, and ‎various obstacles of traditional data centers, ‎mobile edge ‎computing (MEC) has ‎emerged as a promising solution that extends cloud data-processing and decision-making to edge devices. ‎By adopting a MEC structure, IoT applications could be executed ‎locally, on ‎an edge server, different fog nodes, or distant cloud ‎data centers. However, we are ‎often ‎faced with wanting to optimize conflicting criteria such as ‎minimizing energy ‎consumption of limited local capabilities (in terms of CPU, RAM, storage, bandwidth) of mobile edge ‎devices and trying to ‎keep ‎high performance (reducing ‎response time, increasing throughput and service availability) ‎at the same ‎time‎. Achieving one goal may affect the other, making task offloading (TO), ‎resource allocation (RA), and service placement (SP) complex ‎processes. ‎It is a nontrivial multi-objective optimization ‎problem ‎to study the trade-off between conflicting criteria. ‎The paper provides a survey on different TO, SP, and RA recent multi-‎objective optimization (MOO) approaches used in edge computing environments, particularly artificial intelligent (AI) ones, to satisfy various objectives, constraints, and dynamic conditions related to IoT applications‎.

Keywords: mobile edge computing, multi-objective optimization, artificial ‎intelligence ‎approaches, task offloading, resource allocation, ‎ service placement

Procedia PDF Downloads 86

Authors: Mpho Mokoatle, Darlington Mapiye, James Mashiyane, Stephanie Muller, Gciniwe Dlamini

Abstract:

Great advances in high-throughput sequencing technologies have resulted in availability of huge amounts of sequencing data in public and private repositories, enabling a holistic understanding of complex biological phenomena. Sequence data are used for a wide range of applications such as gene annotations, expression studies, personalized treatment and precision medicine. However, this rapid growth in sequence data poses a great challenge which calls for novel data processing and analytic methods, as well as huge computing resources. In this work, a machine and statistical learning approach for DNA sequence classification based on $k$-mer representation of sequence data is proposed. The approach is tested using whole genome sequences of Mycobacterium tuberculosis (MTB) isolates to (i) reduce the size of genomic sequence data, (ii) identify an optimum size of k-mers and utilize it to build classification models, (iii) predict the phenotype from whole genome sequence data of a given bacterial isolate, and (iv) demonstrate computing challenges associated with the analysis of whole genome sequence data in producing interpretable and explainable insights. The classification models were trained on 104 whole genome sequences of MTB isoloates. Cluster analysis showed that k-mers maybe used to discriminate phenotypes and the discrimination becomes more concise as the size of k-mers increase. The best performing classification model had a k-mer size of 10 (longest k-mer) an accuracy, recall, precision, specificity, and Matthews Correlation coeffient of 72.0%, 80.5%, 80.5%, 63.6%, and 0.4 respectively. This study provides a comprehensive approach for resampling whole genome sequencing data, objectively selecting a k-mer size, and performing classification for phenotype prediction. The analysis also highlights the importance of increasing the k-mer size to produce more biological explainable results, which brings to the fore the interplay that exists amongst accuracy, computing resources and explainability of classification results. However, the analysis provides a new way to elucidate genetic information from genomic data, and identify phenotype relationships which are important especially in explaining complex biological mechanisms.

Keywords: AWD-LSTM, bootstrapping, k-mers, next generation sequencing

Procedia PDF Downloads 137

Authors: Darlington Mapiye, Mpho Mokoatle, James Mashiyane, Stephanie Muller, Gciniwe Dlamini

Abstract:

Great advances in high-throughput sequencing technologies have resulted in availability of huge amounts of sequencing data in public and private repositories, enabling a holistic understanding of complex biological phenomena. Sequence data are used for a wide range of applications such as gene annotations, expression studies, personalized treatment and precision medicine. However, this rapid growth in sequence data poses a great challenge which calls for novel data processing and analytic methods, as well as huge computing resources. In this work, a machine and statistical learning approach for DNA sequence classification based on k-mer representation of sequence data is proposed. The approach is tested using whole genome sequences of Mycobacterium tuberculosis (MTB) isolates to (i) reduce the size of genomic sequence data, (ii) identify an optimum size of k-mers and utilize it to build classification models, (iii) predict the phenotype from whole genome sequence data of a given bacterial isolate, and (iv) demonstrate computing challenges associated with the analysis of whole genome sequence data in producing interpretable and explainable insights. The classification models were trained on 104 whole genome sequences of MTB isoloates. Cluster analysis showed that k-mers maybe used to discriminate phenotypes and the discrimination becomes more concise as the size of k-mers increase. The best performing classification model had a k-mer size of 10 (longest k-mer) an accuracy, recall, precision, specificity, and Matthews Correlation coeffient of 72.0 %, 80.5 %, 80.5 %, 63.6 %, and 0.4 respectively. This study provides a comprehensive approach for resampling whole genome sequencing data, objectively selecting a k-mer size, and performing classification for phenotype prediction. The analysis also highlights the importance of increasing the k-mer size to produce more biological explainable results, which brings to the fore the interplay that exists amongst accuracy, computing resources and explainability of classification results. However, the analysis provides a new way to elucidate genetic information from genomic data, and identify phenotype relationships which are important especially in explaining complex biological mechanisms

Keywords: AWD-LSTM, bootstrapping, k-mers, next generation sequencing

Procedia PDF Downloads 126

Authors: Nurullah Ekmekci, Omer Akkaya, Vural Cagliyan

Abstract:

Increased use of the Internet has resulted in the emergence of a new retail types called online shopping or electronic retail (e-retail). The rapid growth of the Internet has enabled customers to search information about the product and buy these products or services from e-retailers. Although this new form of shopping has grown in a remarkable way because of offering easiness to people, it is not an easy task to capture the success by distinguishing from competitors in this environment which millions of players takes place. For the success, e-retailers should determine the factors which the customers take notice while they are buying from e-retailers. This paper aims to identify the factors that provide preferability for the online shopping websites and the importance levels of these factors. These main criteria which have taken notice are Customer Service Performance (CSP), Website Performance (WSP), Criteria Related to Product (CRP), Ease of Payment (EP), Security/Privacy (SP), Ease of Return (ER), Delivery Service Performance (DSP) and Order Fulfillment Performance (OFP). It has benefited from Analytic Hierarchy Process to determine the priority of the criteria. Based on analysis, Security/Privacy (SP) criteria seems to be most important criterion with 22 % weight. Companies should attach importance to the security and privacy for making their online website more preferable among the online shoppers.

Keywords: AHP (analytical hierarchy process), multi-criteria decision making, online shopping, shopping

Procedia PDF Downloads 217

Authors: Pavan K. Rallabandi, Kailash C. Patidar

Abstract:

In this paper, we present an optimization technique or a learning algorithm using the hybrid architecture by combining the most popular sequence recognition models such as Recurrent Neural Networks (RNNs) and Hidden Markov models (HMMs). In order to improve the sequence or pattern recognition/ classification performance by applying a hybrid/neural symbolic approach, a gradient descent learning algorithm is developed using the Real Time Recurrent Learning of Recurrent Neural Network for processing the knowledge represented in trained Hidden Markov Models. The developed hybrid algorithm is implemented on automata theory as a sample test beds and the performance of the designed algorithm is demonstrated and evaluated on learning the deterministic finite state automata.

Keywords: hybrid systems, hidden markov models, recurrent neural networks, deterministic finite state automata

Procedia PDF Downloads 358

Authors: Yavuz Unal, Kemal Polat, H. Erdinc Kocer

Abstract:

In this study, a decision support system comprising three stages has been proposed to detect the disc abnormalities of the lumbar region. In the first stage named the feature extraction, T2-weighted sagittal and axial Magnetic Resonance Images (MRI) were taken from 55 people and then 27 appearance and shape features were acquired from both sagittal and transverse images. In the second stage named the feature weighting process, k-means clustering based feature weighting (KMCBFW) proposed by Gunes et al. Finally, in the third stage named the classification process, the classifier algorithms including multi-layer perceptron (MLP- neural network), support vector machine (SVM), Naïve Bayes, and decision tree have been used to classify whether the subject has lumbar disc or not. In order to test the performance of the proposed method, the classification accuracy (%), sensitivity, specificity, precision, recall, f-measure, kappa value, and computation times have been used. The best hybrid model is the combination of k-means clustering based feature weighting and decision tree in the detecting of lumbar disc disease based on both sagittal and axial MR images.

Keywords: lumbar disc abnormality, lumbar MRI, lumbar spine, hybrid models, hybrid features, k-means clustering based feature weighting

Procedia PDF Downloads 499

Authors: Jairson Barbosa Rodrigues, Paulo Romero Martins Maciel, Germano Crispim Vasconcelos

Abstract:

This paper presents an investigation of the performance impacts regarding the variation of five factors (input data size, node number, cores, memory, and disks) when applying a distributed implementation of Naïve Bayes for text classification of a large Corpus on the Spark big data processing framework. Problem: The algorithm's performance depends on multiple factors, and knowing before-hand the effects of each factor becomes especially critical as hardware is priced by time slice in cloud environments. Objectives: To explain the functional relationship between factors and performance and to develop linear predictor models for time and cost. Methods: the solid statistical principles of Design of Experiments (DoE), particularly the randomized two-level fractional factorial design with replications. This research involved 48 real clusters with different hardware arrangements. The metrics were analyzed using linear models for screening, ranking, and measurement of each factor's impact. Results: Our findings include prediction models and show some non-intuitive results about the small influence of cores and the neutrality of memory and disks on total execution time, and the non-significant impact of data input scale on costs, although notably impacts the execution time.

Keywords: big data, design of experiments, distributed machine learning, natural language processing, spark

Procedia PDF Downloads 86

Authors: Lichung Jen, Yi Chun Liu, Kuan-Wei Lee

Abstract:

Fruitful data has been accumulated in database nowadays and is commonly used as support for decision-making. In the healthcare industry, hospital, for instance, ordering pharmacy inventory is one of the key decision. With large drug inventory, the current cost increases and its expiration dates might lead to future issue, such as drug disposal and recycle. In contrast, underestimating demand of the pharmacy inventory, particularly standing drugs, affects the medical treatment and possibly hospital reputation. Prescription behaviour of hospital physicians is one of the critical factor influencing this decision, particularly irregular prescription behaviour. If a drug’s usage amount in the month is irregular and less than the regular usage, it may cause the trend of subsequent stockpiling. On the contrary, if a drug has been prescribed often than expected, it may result in insufficient inventory. We proposed a hierarchical Bayesian mixture model with two components to identify physicians’ regular/irregular prescription patterns with probabilities. Heterogeneity of hospital is considered in our proposed hierarchical Bayes model. The result suggested that modeling the prescription patterns of physician is beneficial for estimating the order quantity of medication and pharmacy inventory management of the hospital. Managerial implication and future research are discussed.

Keywords: hierarchical Bayesian model, poission mixture model, medicines prescription behavior, irregular behavior

Procedia PDF Downloads 102

Authors: M. Hilmi Ozdemir, Gokhan Ozkan

Abstract:

Performance Based Logistics (PBL) is a strategic approach that enables creating long-term and win-win relations among stakeholders in the acquisition. Contrary to the traditional single transactions, the expected value is created by the performance of the service pertaining to the strategic relationships in this approach. PBL motivates all relevant stakeholders to focus on their core competencies to produce the desired outcome in a collective way. The desired outcome can only be assured with a cost effective way as long as it is periodically measured with the right performance parameters. Thus, defining these parameters is a crucial step for the PBL contracts. In performance parameter determination, Analytic Hierarchy Process (AHP), which is a multi-criteria decision making methodology for complex cases, was used within this study for a complex system. AHP has been extensively applied in various areas including supply chain, inventory management, outsourcing, and logistics. This methodology made it possible to convert end-user’s main operation and maintenance requirements to sub criteria contained by a single performance parameter. Those requirements were categorized and assigned weights by the relevant stakeholders. Single performance parameter capable of measuring the overall performance of a complex system is the major outcome of this study. The parameter deals with the integrated assessment of different functions spanning from training, operation, maintenance, reporting, and documentation that are implemented within a complex system. The aim of this study is to show the methodology and processes implemented to identify a single performance parameter for measuring the whole performance of a complex system within a PBL contract. AHP methodology is recommended as an option for the researches and the practitioners who seek for a lean and integrated approach for performance assessment within PBL contracts. The implementation of AHP methodology in this study may help PBL practitioners from methodological perception and add value to AHP in becoming prevalent.

Keywords: analytic hierarchy process, performance based logistics, performance measurement, performance parameters

Procedia PDF Downloads 259

Authors: Roopa Singh, Anurag Singh, Ajay

Abstract:

Agriculture contributes a major share in the national economy of India. A major portion of Indian economy (about 70%) depends upon agriculture as it forms the main source of income. About 43% of India’s geographical area is used for agricultural activity which involves 65-75% of total population of India. The given work deals with the Fast moving Consumer Goods (FMCG) industries and their inventories which use agricultural produce as their raw material or input for their final product. Since the beginning of inventory practices, many developments took place which can be categorised into three phases, based on the review of various works. The first phase is related with development and utilization of Economic Order Quantity (EOQ) model and methods for optimizing costs and profits. Second phase deals with inventory optimization method, with the purpose of balancing capital investment constraints and service level goals. The third and recent phase has merged inventory control with electrical control theory. Maintenance of inventory is considered negative, as a large amount of capital is blocked especially in mechanical and electrical industries. But the case is different in food processing and agro-based industries and their inventories due to cyclic variation in the cost of raw materials of such industries which is the reason for selection of these industries in the mentioned work. The application of electrical control theory in inventory control makes the decision-making highly instantaneous for FMCG industries without loss in their proposed profits, which happened earlier during first and second phases, mainly due to late implementation of decision. The work also replaces various inventories and work-in-progress (WIP) related errors with their monetary values, so that the decision-making is fully target-oriented.

Keywords: control theory, inventory control, manufacturing sector, EOQ, feedback, FMCG sector

Procedia PDF Downloads 334

Authors: Sanchali Das, Swapan Debbarma

Abstract:

Music mood classification is developing in both the areas of music information retrieval (MIR) and natural language processing (NLP). Some of the Indian languages like Hindi English etc. have considerable exposure in MIR. But research in mood classification in regional language is very less. In this paper, powerful audio based feature for Kokborok Christian song is identified and mood classification task has been performed. Kokborok is an Indo-Burman language especially spoken in the northeastern part of India and also some other countries like Bangladesh, Myanmar etc. For performing audio-based classification task, useful audio features are taken out by jMIR software. There are some standard audio parameters are there for the audio-based task but as known to all that every language has its unique characteristics. So here, the most significant features which are the best fit for the database of Kokborok song is analysed. The regression-based model is used to find out the independent parameters that act as a predictor and predicts the dependencies of parameters and shows how it will impact on overall classification result. For classification WEKA 3.5 is used, and selected parameters create a classification model. And another model is developed by using all the standard audio features that are used by most of the researcher. In this experiment, the essential parameters that are responsible for effective audio based mood classification and parameters that do not significantly change for each of the Christian Kokborok songs are analysed, and a comparison is also shown between the two above model.

Keywords: Christian Kokborok song, mood classification, music information retrieval, regression

Procedia PDF Downloads 192

Authors: W. H. Wan Zakiyatussariroh, A. A. Nasuhar, W. Y. Wan Fairos, Z. A. Nazatul Shahreen

Abstract:

Forecasting dengue incidence in a population can provide useful information to facilitate the planning of the public health intervention. Many studies on dengue cases in Malaysia were conducted but are limited in modeling the outbreak and forecasting incidence. This article attempts to propose the most appropriate time series model to explain the behavior of dengue incidence in Malaysia for the purpose of forecasting future dengue outbreaks. Several seasonal auto-regressive integrated moving average (SARIMA) models were developed to model Malaysia’s number of dengue incidence on weekly data collected from January 2001 to December 2011. SARIMA (2,1,1)(1,1,1)52 model was found to be the most suitable model for Malaysia’s dengue incidence with the least value of Akaike information criteria (AIC) and Bayesian information criteria (BIC) for in-sample fitting. The models further evaluate out-sample forecast accuracy using four different accuracy measures. The results indicate that SARIMA (2,1,1)(1,1,1)52 performed well for both in-sample fitting and out-sample evaluation.

Keywords: time series modeling, Box-Jenkins, SARIMA, forecasting

Procedia PDF Downloads 450

Authors: Sidique Gawusu, Xiaobing Zhang

Abstract:

Turbulence models play a significant role in all computational fluid dynamics based modelling approaches. There is, however, no general turbulence model suitable for all flow scenarios. Therefore, a successful numerical modelling approach is only achievable if a more appropriate closure model is used. This paper evaluates different turbulence models in numerical modelling of oil-water flow within the Eulerian-Eulerian approach. A comparison among the obtained numerical results and published benchmark data showed reasonable agreement. The domain was meshed using structured mesh, and grid test was performed to ascertain grid independence. The evaluation of the models was made through analysis of velocity and pressure profiles across the domain. The models were tested for their suitability to accurately obtain a scalable and precise numerical experience. As a result, it is found that all the models except Standard-ω provide comparable results. The study also revealed new insights on flow in porous media, specifically oil reservoirs.

Keywords: turbulence modelling, simulation, multi-phase flows, water-flooding, heavy oil

Procedia PDF Downloads 252

Authors: Teklu Urgessa, Wookjae Maeng, Joong Seek Lee

Abstract:

Application of five implementations of three data mining classification techniques was experimented for extracting important insights from tourism data. The aim was to find out the best performing algorithm among the compared ones for tourism knowledge discovery. Knowledge discovery process from data was used as a process model. 10-fold cross validation method is used for testing purpose. Various data preprocessing activities were performed to get the final dataset for model building. Classification models of the selected algorithms were built with different scenarios on the preprocessed dataset. The outperformed algorithm tourism dataset was Random Forest (76%) before applying information gain based attribute selection and J48 (C4.5) (75%) after selection of top relevant attributes to the class (target) attribute. In terms of time for model building, attribute selection improves the efficiency of all algorithms. Artificial Neural Network (multilayer perceptron) showed the highest improvement (90%). The rules extracted from the decision tree model are presented, which showed intricate, non-trivial knowledge/insight that would otherwise not be discovered by simple statistical analysis with mediocre accuracy of the machine using classification algorithms.

Keywords: classification algorithms, data mining, knowledge discovery, tourism

Procedia PDF Downloads 269

Authors: Thanh Nguyen, Andrei Doncescu, Pierre Siegel

Abstract:

Classification is an important data mining technique and could be used as data filtering in artificial intelligence. The broad application of classification for all kind of data leads to be used in nearly every field of our modern life. Classification helps us to put together different items according to the feature items decided as interesting and useful. In this paper, we compare two classification methods Naïve Bayes and ADTree use to detect spam e-mail. This choice is motivated by the fact that Naive Bayes algorithm is based on probability calculus while ADTree algorithm is based on decision tree. The parameter settings of the above classifiers use the maximization of true positive rate and minimization of false positive rate. The experiment results present classification accuracy and cost analysis in view of optimal classifier choice for Spam Detection. It is point out the number of attributes to obtain a tradeoff between number of them and the classification accuracy.

Keywords: classification, data mining, spam filtering, naive bayes, decision tree

Procedia PDF Downloads 388

Authors: Mustafa Alhamdi

Abstract:

Industrial application to classify gamma rays and neutron events is investigated in this study using deep machine learning. The identification using a convolutional neural network and recursive neural network showed a significant improvement in predication accuracy in a variety of applications. The ability to identify the isotope type and activity from spectral information depends on feature extraction methods, followed by classification. The features extracted from the spectrum profiles try to find patterns and relationships to present the actual spectrum energy in low dimensional space. Increasing the level of separation between classes in feature space improves the possibility to enhance classification accuracy. The nonlinear nature to extract features by neural network contains a variety of transformation and mathematical optimization, while principal component analysis depends on linear transformations to extract features and subsequently improve the classification accuracy. In this paper, the isotope spectrum information has been preprocessed by finding the frequencies components relative to time and using them as a training dataset. Fourier transform implementation to extract frequencies component has been optimized by a suitable windowing function. Training and validation samples of different isotope profiles interacted with CdTe crystal have been simulated using Geant4. The readout electronic noise has been simulated by optimizing the mean and variance of normal distribution. Ensemble learning by combing voting of many models managed to improve the classification accuracy of neural networks. The ability to discriminate gamma and neutron events in a single predication approach using deep machine learning has shown high accuracy using deep learning. The paper findings show the ability to improve the classification accuracy by applying the spectrogram preprocessing stage to the gamma and neutron spectrums of different isotopes. Tuning deep machine learning models by hyperparameter optimization of neural network models enhanced the separation in the latent space and provided the ability to extend the number of detected isotopes in the training database. Ensemble learning contributed significantly to improve the final prediction.

Keywords: machine learning, nuclear physics, Monte Carlo simulation, noise estimation, feature extraction, classification

Procedia PDF Downloads 122

Authors: Ali Akber Jaffri, Asad Naseem, Javeria Khan, Zubair Hamza, Ishtiaq

Abstract:

The purpose of this study is to determine safety stock, maximum inventory level, reordering point, and reordering quantity by rearranging lot sizes for supplier and customer in MRO (maintenance repair operations) warehouse of Engro Polymers & Chemicals. To achieve the aim, physical analysis method and excel commands were carried out to elicit the customer and supplier data provided by the company. Initially, we rearranged the current lot sizes and MOUs (measure of units) in SAP software. Due to change in lot sizes, we have to determine the new quantities for safety stock, maximum inventory, reordering point and reordering quantity as per company's demand. By proposed system, we saved extra cost in terms of reducing time of receiving from vendor and in issuance to customer, ease of material handling in MRO warehouse and also reduce human efforts.

Keywords: maintenance repair operation, maximum inventory, reorder quantity, safety stock

Procedia PDF Downloads 253

Authors: Mohammad Sarchami, Mohsen Zeinalkhani

Abstract:

Always, financial reporting system faces some problems to win public ear. The increase in the number of fraud and representation, often combined with the bankruptcy of large companies, has raised concerns about the quality of financial statements. So, investors, legislators, managers, and auditors have focused on significant fraud detection or prevention in financial statements. This article aims to investigate the Sugeno fuzzy classification to consider fraud detection in financial reporting of accepted firms by Tehran stock exchange. The hypothesis is: Sugeno fuzzy classification may detect fraud in financial reporting by financial ratio. Hypothesis was tested using Matlab software. Accuracy average was 81/80 in Sugeno fuzzy classification; so the hypothesis was confirmed.

Keywords: fraud, financial reporting, Sugeno fuzzy classification, firm

Procedia PDF Downloads 221

Authors: Victoria Molina, Wendy Franco, Sergio Benavides, José M. Troncoso, Ricardo Luna, Jose R. PéRez-Correa

Abstract:

Emamectin Benzoate (EB) is a crystal antiparasitic that belongs to the avermectin family. It is one of the most common treatments used in Chile to control Caligus rogercresseyi in Atlantic salmon. However, the sea lice acquired resistance to EB when it is exposed at sublethal EB doses. The low solubility rate of EB and its degradation at the acidic pH in the fish digestive tract are the causes of the slow absorption of EB in the intestine. To protect EB from degradation and enhance its absorption, specific microencapsulation technologies must be developed. Amorphous Solid Dispersion techniques such as Spray Drying (SD) and Ionic Gelation (IG) seem adequate for this purpose. Recently, Soluplus® (SOL) has been used to increase the solubility rate of several drugs with similar characteristics than EB. In addition, alginate (ALG) is a widely used polymer in IG for biomedical applications. Regardless of the encapsulation technique, the quality of the obtained microparticles is evaluated with the following responses, yield (Y%), encapsulation efficiency (EE%) and loading capacity (LC%). In addition, it is important to know the percentage of EB released from the microparticles in gastric (GD%) and intestinal (ID%) digestions. In this work, we microencapsulated EB with SOL (EB-SD) and with ALG (EB-IG) using SD and IG, respectively. Quality microencapsulation responses and in vitro gastric and intestinal digestions at pH 3.35 and 7.8, respectively, were obtained. A central composite design was used to find the optimum microencapsulation variables (amount of EB, amount of polymer and feed flow). In each formulation, the behavior of these variables was predicted with statistical models. Then, the response surface methodology was used to find the best combination of the factors that allowed a lower EB release in gastric conditions, while permitting a major release at intestinal digestion. Two approaches were used to determine this. The desirability approach (DA) and multi-objective optimization (MOO) with multi-criteria decision making (MCDM). Both microencapsulation techniques allowed to maintain the integrity of EB in acid pH, given the small amount of EB released in gastric medium, while EB-IG microparticles showed greater EB release at intestinal digestion. For EB-SD, optimal conditions obtained with MOO plus MCDM yielded a good compromise among the microencapsulation responses. In addition, using these conditions, it is possible to reduce microparticles costs due to the reduction of 60% of BE regard the optimal BE proposed by (DA). For EB-GI, the optimization techniques used (DA and MOO) yielded solutions with different advantages and limitations. Applying DA costs can be reduced 21%, while Y, GD and ID showed 9.5%, 84.8% and 2.6% lower values than the best condition. In turn, MOO yielded better microencapsulation responses, but at a higher cost. Overall, EB-SD with operating conditions selected by MOO seems the best option, since a good compromise between costs and encapsulation responses was obtained.

Keywords: microencapsulation, multiple decision-making criteria, multi-objective optimization, Soluplus®

Procedia PDF Downloads 100