Applications of Mathematical Modeling, Machine Learning, and Intelligent Computing for Industrial Development

Description

The text focuses on mathematical modeling and applications of advanced techniques of machine learning, and artificial intelligence, including artificial neural networks, evolutionary computing, data mining, and fuzzy systems to solve performance and design issues more precisely. Intelligent computing encompasses technologies, algorithms, and models in providing effective and efficient solutions to a wide range of problems, including the airport’s intelligent safety system. It will serve as an ideal reference text for senior undergraduate, graduate students, and academic researchers in fields that include industrial engineering, manufacturing engineering, computer engineering, and mathematics. The book: Discusses mathematical modeling for traffic, sustainable supply chain, vehicular Ad-Hoc networks, and internet of things networks with intelligent gatewaysCovers advanced machine learning, artificial intelligence, fuzzy systems, evolutionary computing, and data mining techniques for real- world problemsPresents applications of mathematical models in chronic diseases such as kidney and coronary artery diseasesHighlights advances in mathematical modeling, strength, and benefits of machine learning and artificial intelligence, including driving goals, applicability, algorithms, and processes involvedShowcases emerging real-life topics on mathematical models, machine learning, and intelligent computing using an interdisciplinary approach The text presents emerging real-life topics on mathematical models, machine learning, and intelligent computing in a single volume. It will serve as an ideal text for senior undergraduate students, graduate students, and researchers in diverse fields, including industrial and manufacturing engineering, computer engineering, and mathematics. Cover Half Title Series Page Title Page Copyright Page Table of Contents Preface Editors’ biographies List of contributors Section 1: Mathematical modeling Chapter 1: An interactive weight-based portfolio system using goal programming 1.1 Introduction 1.2 Interactive weight-based portfolio system 1.3 Stock data 1.4 Experimental work 1.5 Result analysis 1.6 Comparative analysis 1.7 Prototype of interactive weight-based portfolio system 1.8 Conclusion References Chapter 2: Generalized knowledge measure based on hesitant fuzzy sets with application to MCDM 2.1 Introduction 2.2 Fuzzy sets and hesitant fuzzy sets 2.3 Projection measure of HFEs 2.4 Entropy and knowledge measure 2.4.1 Applications of entropy and knowledge measures 2.5 Generalization of information measures 2.6 Comparative study 2.7 MCDM 2.8 Problem Statement 2.9 Conclusion References Chapter 3: Queuing system with customers’ impatience, retention, and feedback 3.1 Introduction 3.2 Assumptions of the model 3.3 Formulation of mathematical model 3.4 Solution of the model in steady-state 3.5 Measures of performance 3.5.1 Expected system size (Ls) 3.5.2 Expected waiting time of the customer in the system 3.5.3 Expected waiting time of a customer in the queue 3.5.4 Expected queue length Lq = λWq 3.6 Particular cases 3.7 Numerical illustration and sensitivity analysis 3.8 Conclusion and future work References Chapter 4: Controllable multiprocessor queueing system 4.1 Introduction 4.2 Formulation 4.3 Queue size distribution and execution indices 4.4 Estimated waiting time 4.5 Numerical results 4.6 Discussion References Chapter 5: Multi-server queuing system with feedback and retention 5.1 Introduction 5.2 Queuing model 5.3 Mathematical formulation 5.4 Time dependent solution of the model 5.5 Particular cases 5.6 Conclusion References Chapter 6: Retrial queueing system subject to differentiated vacations with impatient customers 6.1 Introduction 6.2 Model description 6.3 Steady-state solutions 6.4 Performance measures 6.5 Numerical results 6.6 Conclusions and future works References Chapter 7: A Preorder discounts and online payment facility on a sustainable inventory model with controllable carbon emission 7.1 Introduction 7.2 Review of literature 7.2.1 Literature review on inventory model with controllable carbon emission 7.2.2 Literature review on inventory model with non-instantaneous deteriorating items 7.2.3 Literature review of inventory model with the preorder program and multiple discount facility 7.2.4 Literature review of inventory model with inflation 7.2.5 Research gap 7.2.6 Problem definition 7.3 Assumptions and notations 7.3.1 Assumptions 7.4 Mathematical treatment 7.5 Algorithm 7.6 Numerical illustration 7.6.1 Example 1 when E ≥ a 7.6.2 Example 2 when E ≥ a 7.6.3 Example 3 when E < a 7.6.4 Example 4 when E < a 7.7 Result summary 7.8 Concavity 7.9 Sensitivity analysis 7.10 Observations 7.10.1 For case 1 7.10.2 For case 2 7.11 Managerial insights 7.12 Conclusion References Chapter 8: Green inventory systems with two-warehouses for rubber waste items 8.1 Introduction 8.1.1 Concept of two-warehouse in rubber waste items green inventory systems 8.1.2 Genetic algorithm 8.2 Related work 8.3 Notations and assumptions 8.4 Formulation and solution of the rubber waste items green inventory 8.5 Numerical simulations 8.5.1 Implementation of GA 8.6 Conclusions References Section 2: Machine learning Chapter 9: Cyber-attack detection applying machine learning approach 9.1 Introduction 9.2 Known attack detection methods 9.3 Unknown attack 9.3.1 Anomaly-based approaches to defending zero-day cyber-attacks 9.3.2 Graph-based approaches to defending zero-day cyber-attacks 9.3.3 ML and deep learning-based approaches to defend against unknown (zero-day) cyber-attacks 9.4 Feature reduction and data generation for intrusion detection 9.4.1 Feature reduction techniques for attack data 9.4.2 Data generation methods to handle class imbalance problem 9.5 Conclusion References Chapter 10: Feature extraction methods for intelligent audio signal classification 10.1 Introduction 10.2 Theoretical background 10.3 Deployment of the three methods 10.3.1 Sound event detection 10.4 Performance Measures Metrics (PMMs) 10.4.1 Quantification of evaluation criteria 10.4.2 PMMs criteria for evaluation and selection 10.4.3 Manual microphones calibration process 10.5 Conclusions References Chapter 11: Feature detection and extraction techniques with different similarity measures using bag of features scheme 11.1 Introduction 11.2 Theoretical background 11.3 Deployment of the method 11.3.1 Data set 11.3.2 Proposed methodology 11.3.3 Proposed algorithm 11.3.4 MATLAB implementation 11.4 Results 11.5 Performance measures 11.5.1 Comparison of different methods descriptors 11.5.2 Evaluation 11.5.3 Performance compared with different algorithms 11.6 Conclusions References Chapter 12: Stratification based initialization model on partitional clustering for big data mining 12.1 Introduction 12.2 Partitional based clustering approach for big data 12.3 Clustering techniques for big data 12.3.1 Incremental method 12.3.2 Divide and conquer method 12.3.3 Data summarization 12.3.4 Sampling-based methods 12.3.5 Efficient Nearest Neighbor (NN) Search 12.3.6 Dimension reduction-based techniques 12.3.7 Parallel computing methods 12.3.8 Condensation-based methods 12.3.9 Granular computing 12.4 Sampling methods for big data mining 12.4.1 Uniform random sampling 12.4.2 Systematic sampling 12.4.3 Progressive sampling 12.4.4 Reservoir sampling 12.4.5 Stratified sampling 12.5 Proposed model for partitional clustering algorithms using stratified sampling 12.5.1 Stratified sampling 12.5.2 Stratification 12.5.3 Proposed initialization model for partitional based clustering 12.5.4 Experimental environment and selected algorithm 12.5.5 Validation criteria 12.5.6 Results and discussion 12.6 Conclusion References Chapter 13: Regression ensemble techniques with technical indicators for prediction of financial time series data 13.1 Introduction 13.2 Literature review 13.3 Dataset and technical indicators 13.4 Framework of proposed work 13.5 Methodology 13.5.1 Regression 13.5.2 Ensemble model 13.5.3 K-Fold cross validation 13.6 Result and analysis 13.7 Conclusion References Chapter 14: Intelligent system for integrating customer’s voice with CAD for seat comfort 14.1 Introduction 14.2 Quality Function Deployment (QFD) 14.2.1 QFD design parameters (HOWs) 14.2.2 Comfort evaluation using QFD 14.3 CAD for seat comfort assessment 14.3.1 The seat models 14.3.2 Seat features 14.3.3 Cushions materials 14.3.4 The human model 14.3.5 Human’s material properties 14.3.6 Occupant anthropometric analysis 14.3.7 The simulation technique 14.3.8 Finite Element Analysis (FEA) 14.4 The prediction model 14.4.1 Machine learning predictive model 14.5 System validation and calibration 14.6 Results and discussions 14.7 Conclusions References Chapter 15: An implementation of machine learning to detect real time web application vulnerabilities 15.1 Introduction 15.2 Architecture of vulnerability scanners 15.2.1 Objective 15.2.2 Techniques for vulnerability scanning 15.3 Recognition of vulnerabilities automatically 15.3.1 Crawling component 15.3.2 Component of attack 15.3.3 Analysis modules 15.4 Principles of attack and analyzation 15.4.1 SQL injection attack 15.4.2 Simple reflected XSS attack 15.4.3 Encoded reflected XSS 15.4.4 Form-redirecting XSS 15.5 Outcome 15.6 Estimated consequences References Section 3: Intelligent computing Chapter 16: Intelligent decision support system for air traffic management 16.1 Introduction 16.2 Theory and technical approach 16.3 Experimental work 16.4 Discussion 16.5 Conclusion and future work References Chapter 17: Bio-mechanical hand with wide degree of freedom 17.1 Introduction 17.2 Theoretical background 17.2.1 Steps to extract EMG signals 17.2.2 Types of designs 17.2.3 Problem formulation 17.2.4 Proper fitting 17.2.5 Delayed impulse response 17.2.6 Recent discovered materials 17.3 Electronics circuits 17.4 Hardware designing 17.4.1 Hardware used 17.4.2 Software used 17.4.3 Designing steps 17.4.4 Image capturing by fyuse 17.4.5 File format conversion in blender 17.4.6 Flattening design in solid works 17.4.7 Conversion to mesh file 17.4.8 Conversion to g-code (Geometric codes) 17.4.9 3D printing 17.4.10 Printed hand 17.4.11 Circuit outline 17.4.12 Pseudo code/Algorithm for Arduino 17.5 Result and discussion 17.5.1 Appearance 17.6 Conclusion and future scope 17.6.1 Conclusion 17.6.2 Future scope References Chapter 18: Analyses of repetitive motions in goods to person order picking strategies using intelligent human factors system 18.1 Introduction 18.1.1 Background 18.1.2 Goods to person versus person to goods 18.1.3 Goods to persons repetitive motion’s effects on human pickers 18.2 Methodology 18.2.1 The digital pickers 18.2.2 Warehouse layout components 18.2.3 Repetitive task analyses 18.3 Data collection and results 18.3.1 Comfort assessment 18.3.2 Fatigue and recovery analysis 18.3.3 Lower Back Analysis (LBA) 18.3.4 Metabolic Energy Expenditure (MEE) 18.3.5 Static Strength Prediction (SSP) 18.3.6 Rapid Upper Limb Assessment (RULA) and Ovako working posture analysis 18.3.7 Ergonomic results for Eva (50th percentile) 18.3.8 Ergonomic results for Bill (50th percentile) 18.3.9 Ergonomic results for Jill (5th percentile) 18.4 Discussion 18.5 Conclusion References Chapter 19: Hybridization of IoT networks with intelligent gateways 19.1 Introduction 19.2 Basic IoT architecture 19.3 Proposed intelligent gateway-based hybridized IoT network 19.4 Machine learning algorithms for proposed model 19.4.1 k-Nearest Neighbor (kNN) algorithm 19.4.2 SVM (Support Vector Machine) 19.5 Comparative analysis 19.5.1 Accuracy 19.5.2 Receiver Operating Characteristic (ROC) 19.6 Transmission reliability of proposed model 19.7 Conclusion References Chapter 20: M/M1 + M2/1/N/K-policy queues to monitor waiting times via intelligent computing 20.1 Introduction 20.1.1 False alarm rate for control chart and average run length (ARL) 20.1.2 Applications of queueing models 20.1.3 Examples of queueing systems 20.1.4 Statistical Process Control (SPC) 20.2 Steady state distributions of M/M1 + M2 /1/N/K-policy queues 20.2.1 Moments of waiting time process and control limits 20.3 Sampling plan and simulation 20.3.1 Embedded Markov chains 20.3.2 Simulation study 20.4 Application: find an optimum K-policy 20.5 Discussion, conclusion, and scope References Index