Methods and Applications of Artificial Intelligence

Abdelkhalak El Hami is a university professor at INSA Rouen Normandie, France. He is the author/co-author of over sixty books and is responsible for several European educational and/or research projects. He is a specialist in the optimization, reliability and AI of multiphysical systems.

• Preface xiAbdelkhalak EL HAMIChapter 1. Dynamic of PP-GF70 Using an Intelligent Method 1Abdelilah ELBAZZE and Bouchaib RADI1.1. Introduction 11.2. Crash analysis 31.2.1. Dynamic equations 31.2.2. Explicit simulation 71.2.3. Shell elements 111.3. Application to automobile crashes 121.3.1. Description of the study 121.3.2. Model preprocessing 131.3.3. Mesh generation according to ANSA 141.3.4. Materials and properties 141.3.5. Generation of part assembly connections 151.3.6. Assembly of components 161.3.7. Interfaces 171.3.8. Side crash with mobile barrier 181.4. Numerical results and discussions 211.4.1. Numerical stability of the model 211.4.2. Model behavior 231.5. Behavior of the target part during impact 241.5.1. Deformation of the target part 241.5.2. Plastic deformation 261.6. Conclusion 271.7. References 27Chapter 2. AI Integration into Additive Manufacturing 31Adnane ZOUBEIR, Bouchaib RADI and Abdelkhalak EL HAMI2.1. Introduction 312.2. Additive manufacturing processes 322.2.1. Principle of additive manufacturing 322.2.2. Various stages of additive manufacturing 332.3. Integration of AI into additive manufacturing 352.3.1. Introduction 352.3.2. Feasibility of AI-based solutions 362.3.3. Training data 362.3.4. Prediction models and selection of the proper solution 372.4. Development of an AI model for additive manufacturing 372.4.1. Introduction 372.4.2. Machine learning and deep learning 382.4.3. Data preparation for training an AI model 382.4.4. AI model training 392.4.5. Performance of the AI model prediction 412.4.6. Application to the optimization of manufacturing parameters 412.5. Conclusion 432.6. References 44Chapter 3. Optimization of BGA Components under Real Service Conditions 47Sinda GHENAM, Abdelkhalak EL HAMI, Wajih GAFSI, Ali AKROUT and Mohamed HADDAR3.1. Introduction 473.2. Presentation of the optimization method 493.3. Presentation of the system 503.4. Methodology 513.4.1. Deterministic design optimization (DDO) 513.4.2. Reliability-based design optimization (RBDO) 523.5. Results 553.5.1. DDO 553.5.2. RBDO 573.6. Conclusion 603.7. References 61Chapter 4. Electronic Toll Collection in the Age of Connected Vehicles 65Adnane CABANI and Houcine CHAFOUK4.1. Introduction 654.2. Electronic toll collection service using ITS-G5 664.2.1. Constant velocity (CV) model 694.2.2. Constant acceleration (CA) model 704.2.3. Constant turn rate and velocity (CTRV) model 704.2.4. Constant turn rate and acceleration (CTRA) model 714.3. Experimental results of the path tracking 724.4. Conclusion 744.5. References 74Chapter 5. At the Core of Artificial Intelligence: Leveraging Machine Learning through Random Forest 77Anas EL ATTAOUI, Younes KOULOU and Norelislam EL HAMI5.1. Overview of AI and its branches 785.1.1. AI and machine learning 785.1.2. Robust techniques are required for classification and regression 795.1.3. Introduction to the random forest method 795.2. Decision trees 795.2.1. Structure of a decision tree and associated terms 795.2.2. Entropy 825.2.3. Information gain 835.2.4. Gini index 835.2.5. Decision tree generation algorithm 845.2.6. Example of decision tree construction 865.2.7. Advantages and limitations of decision trees 935.3. Random forest foundations 935.3.1. Origin and basic principle of the random forest 935.3.2. Stages of random forest algorithm 945.3.3. Random forest example for classification. 965.3.4. Random forest example for regression 1025.4. Case study: detection of malware on Android 1105.4.1. Data presentation 1105.4.2. Results 1115.5. Chapter essentials 1135.6. References 114Chapter 6. Linear Regression and Application to Artificial Intelligence 121Sara RHOUAS, Norelislam EL HAMI and Younes KOULOU6.1. Introduction 1216.2. Linear regression model 1226.2.1. Objectives and use 1226.2.2. Applications to everyday life 1236.2.3. Simple model vs. multiple model 1246.2.4. Regression coefficients 1256.3. Methods for assessing coefficients 1266.3.1. Ordinary least squares (OLS) method 1266.3.2. Ridge regression 1276.3.3. LASSO regression 1286.3.4. Elastic Net regression 1296.3.5. Weighted least squares 1316.3.6. Linear regression by gradient descent 1326.4. Model evaluation 1336.4.1. Performance evaluation metrics 1336.4.2. Cross-validation 1366.4.3. Model diagnosis 1396.5. Practical cases 1416.5.1. Data preparation and analysis 1416.5.2. Model training 1476.6. Conclusion 1516.7. References 152Chapter 7. Machine Learning and Artificial Intelligence with XGBoost Algorithm for Binary Classification 161Hakima REDDAD, Maria ZEMZAMI, Norelislam EL HAMI and Nabil HMINA7.1. Introduction 1617.2. XGBoost algorithm: state of the art 1647.2.1. Anatomy of XGBoost algorithm 1657.2.2. Main features of XGBoost algorithm 1667.2.3. XGBoost operation 1737.2.4. Theoretical basis of XGBoost algorithm 1747.2.5. Feature importance 1767.3. Case study: application of the XGBoost algorithm - resolution of a binary classification problem 1777.3.1. Problem statement 1777.3.2. Data preparation 1787.3.3. Exploratory data analysis 1797.3.4. Construction of the predictive model with XGBoost algorithm 1807.3.5. Evaluation of the predictive model 1817.4. Results and discussions 1837.5. Conclusion 1857.6. References 185Chapter 8. Application of Interoperability to Intelligent System of Systems 193Younes KOULOU, Norelislam EL HAMI, Anas EL ATTAOUI and Sara RHOUAS8.1. General context of SoS 1948.1.1. Definitions 1948.1.2. SoS taxonomy 1958.1.3. Properties of SoS 1978.2. Concept of intelligent SoS 2008.2.1. Intelligence in SoS 2008.2.2. Architecture of intelligent SoS 2018.3. Interoperability 2038.3.1. Classification of interoperability. 2048.4. Evaluation of interoperability as a basis for the structural analysis of intelligent SoS 2068.4.1. Quantification of interoperability barriers 2078.4.2. Inefficiency of communication 2098.4.3. Application to the case study 2128.5. Conclusion 2158.6. References 216List of Authors 223Index 225