Artificial Neural Network for Solving the Inventory Control Problem in Fuzzy Environments

In this chapter, different inventory control problems are formulated in fuzzy environment and solved by artificial neural network. Due to present the non-linearity associated with the differential equation in fuzzy environment, the solution procedure may be very complicated. To avoid the situation, artificial neural networks play an important role. In this chapter, different inventory control problems are formulated in fuzzy environment and solved by artificial neural network. Due to present the non-linearity associated with the differential equation in fuzzy environment, the solution procedure may be very complicated. To avoid the situation, artificial neural networks play an important role.

Application of Artificial Neural Networks to Estimate Tensile Strength of Austenitic Stainless Steel During Metal Inert Gas Welding Process

In the present work, artificial neural networks (ANN) have been used to model the complex relationship between input-output parameters of metal inert gas (MIG) welding processes. Four ANN training algorithms such as back propagation neural network (BPNN) with gradient descent momentum (GDM), BPNN with Levenberg Marquardt (LM) algorithm, BPNN with Bayesian regularization (BR), and radial basis function networks (RBFN) method have been used for prediction modelling. An experimentation based on full factorial experimental design has been conducted on MIG welding of austenitic stainless steel of grade-304 where welding current, welding speed, and voltage have been considered as input parameters, and tensile strength has been considered as measurable output parameter. The dataset so constituted is used for ANN modelling. Altogether, 40 different ANN architectures have been trained and tested using the above-mentioned algorithms, and 3-11-1 ANN architecture trained using BPNN with BR has been considered to show best prediction capability with mean % absolute error of 0.354%.

Biological Neural Circuits as Applications in Business and Engineering as a New Approach in Artificial Intelligence

The chapter describes the new approach in artificial intelligence based on simulated biological neurons and created neural circuits which represent the next generation of computing systems and artificial intelligence for business applications. Unlike existing technical devices for implementing a neuron based on classical nodes oriented to binary processing, the proposed path is based on bit-parallel processing of numerical data (synapses) for obtaining result. The proposed approach of implementation a neuron can serve as a new elementary basis for the construction of neuron-based computers with a higher processing speed of biological information and good survivability. The research demonstrates the developed nervous circuit constructor and its usage in building of the nervous circuits of biological creatures and simulation of their work and how it could be used in the next generation of the computing systems.

Comprehensive Introduction to Neural Networks

Unknowingly, artificial intelligence (AI) has become an inevitable part of our lives. In this chapter, the authors discuss how the neural networks, a sub-part of AI, changed the way we analyse things. In this chapter, the advent of neural networks, inspiration from the human brain, simplification models of biological neuron models are discussed. Later, a detailed overview of various neural network models, their strengths, limitations, applications, and challenges are presented in detail.

Nonlinear Filtering in Artificial Neural Network Applications in Business and Engineering

Within this chapter the author considers the possibility of applying modern IT technologies to implement information processing algorithms in the sphere of UAV motion control system. Filtration of coordinates and motion parameters of objects in the situation of uncertainty is carried out using nonlinear adaptive filters, such as: Kalman and Bayesian filters. The author considers numerical methods for digital implementation of nonlinear filters based on the convolution of functions, the possibilities of neural networks, and fuzzy logic for solving the problems of tracking UAV objects (or missiles), the math model of dynamics, the features of the practical implementation of state estimation algorithms in the frame of added additional degrees of freedom. The considered algorithms are oriented on solving the problems in real time using parallel and cloud computing.

A Particle Swarm Optimization-Based Approach for Finding Reliability in a Total Hip Prosthesis

Nowadays, we assist the global extension of reliability optimization problems from the design phase of systems and sub-systems to the design and operational phases, not only of systems and sub-systems, but also of bio functionality design. This chapter investigates the relative performances of particle swarm optimization (PSO) variants when used to find reliability in the total hip prosthesis by finding the maximization of jumping distance (JD) to avoid dislocation and the minimization of system's stability to offer mobility. Statistical analysis of different cases of head diameters of 22, 28, 36, 40 mm has been conducted to survey the convergence and relative performances of the main PSO variants when applied to solve reliability in the total hip prosthesis.

Integrated Kinematic Machining Error Compensation for Impeller Rough Tool Paths Programming in a Step-Nc Format Using Neural Network Approach Prediction

The most important components used in aerospace, ships, and automobiles are designed with free form surfaces. An impeller is one of the most important components that are difficult to machine because of its twisted blades. This research book is based on the premise that a STEP-NC program can document “generic” manufacturing information for an impeller. This way, a STEP-NC program can be made machine-independent and has an advantage over the conventional G-code-based NC program that is always generated for a specific CNC machine. Rough machining is recognized as the most crucial procedure influencing machining efficiency and is critical for the finishing process. The research work reported in this chapter focuses on introduces a fully STEP-compliant CNC by putting forward an interpolation algorithm for non uniform rational basic spline (NURBS) curve system for rough milling tool paths with an aim to solve the problems of kinematic errors solutions in five axis machine by neural network implementation.

Filter Selection for Speaker Diarization Using Homomorphism

In the current scenario, audio classification and recognition of a particular source of voice is a major challenge when several speakers are speaking at a time. The process of identifying each speaker in an audio segment is called speaker diarization. The major steps involved in speaker diarization are speech detection, speaker change, and speaker merges. Finding and suggesting the best filter is one of the most important task involved in every step of this process. No current researches at present have yet focused on the impact of filter with optimized approaches. In this chapter, a simple yet effective method using homorphism has been implemented to recommend the best filter for any audio classification task for this purpose. After having performed the classification task for a number of commonly used filters, their accuracies have been compared at every step of speaker diarization. In this work, a module using Bayesian information criterion, convolutional neural networks is implemented, and diarization algorithm that performs the task of speaker diarization is proposed.

Deep Learning

Deep learning (DL) is a rising star of machine learning (ML) and artificial intelligence (AI) domains. Until 2006, many researchers had attempted to build deep neural networks (DNN), but most of them failed. In 2006, it was proven that deep neural networks are one of the most crucial inventions for the 21st century. Nowadays, DNN are being used as a key technology for many different domains: self-driven vehicles, smart cities, security, automated machines. In this chapter, brief information about DL theory is given, advantages and disadvantages of deep learning are discussed, most used types of DNN are mentioned, popular DL architectures and frameworks are glanced and aimed to build smart systems for the finance and real estate domains. Finally, a case study about image recognition using transfer learning is developed.

SEM

The aim of this chapter is to test the hypothesis that the two-step structural equation modelling (SEM) and artificial neural network (ANN) approach enables better in-depth research results as compared to the single-step SEM approach. This approach was used to determine which factors have statistically significant influence on customer satisfaction and customer loyalty in online shopping. The purpose of this chapter is to extend the role of e-service quality and e-recovery research which is traditionally based on SEM technique with ANN approach. In the first step of the present research, the SEM technique was used to determine which factors have statistically significant influence on customer satisfaction; in the second step, ANN models were used to rank the relative influence of significant predictors obtained from SEM. The results indicate that effectiveness of information content, hedonic shopping value, information security and confidentiality, responsiveness, and website entertainment have a positive impact on customer satisfaction.