Data-Driven Design Optimization for Industrial Products

Abstract

This book describes different approaches for solving industrial problems like product design, process optimization, quality enhancement, productivity improvement and cost minimization. Several optimization techniques are described. The book covers case studies on the applications of classical as well as evolutionary and swarm optimization tools for solving industrial issues. The content is very helpful for industry personnel, particularly engineers from the Operation, R&D and Quality Assurance sectors, and also the academic researchers of different engineering and/or business administration background. Optimization is a standard activity in everyday life. We use it consciously or unconsciously for almost all our daily jobs. When we try to find a way to finish an assignment in our professional life in less effort or time, it is an optimization procedure. Deciding the best way to reach the workplace from the time and distance aspects, depending on the time of the day, is also an optimization. Many people in this world try to reduce the daily cost of living for their survival. All these actions have some amount of mental calculations behind it, and most of the time the calculation is not that mathematical. The optimization process basically finds the values of the variables those control the objective we need to optimize (i.e., min- imize or maximize) while satisfying some constraints. This process becomes mathematical when we employ it in the professional sectors like finance, con- struction, manufacturing, etc. In those cases, the number of variables is quite high, and they are correlated in a complex way. Mathematical optimization has various components. The first is the objective function, which defines the attribute to be optimized in terms of the dependent variables or design variables. For example, in manufacturing process, it describes the profit or the cost or the product quality. The design variables are the variables which control the value of the attribute, which is being optimized. The amounts of different resources used and the time spent in a manufacturing process may be considered as the design variables. The third important component in an opti- mization process is the constraint. It may be single or a set of constraints, which allow the process to take on certain values of the design variables but exclude others. Among the different approaches of optimization, the classical derivative-based approaches were the most established and common optimization methods. But in recent years, advent of metaheuristic methods has changed the domain of opti- mization and made those methods more acceptable due to their ability to handle complex problems and lesser possibility of getting stuck in the local optima. A metaheuristic method has the strategy to guide and modify the pure heuristics to produce better solutions which are beyond the solutions generated in heuristic processes. The metaheuristic optimization algorithms consist of the evolutionary algorithms, the swarm intelligence, other bio- and nature-inspired algorithms, and some Physics and process-based algorithms. In this book, most of the chapters deal with several case studies related to application of these metaheuristic algorithms in different domains of industry. Optimizations methods are extensively being applied to different sectors of the industries and professional life, which includes the information system, the financial operations, the manufacturing systems, engineering design and design optimization, operations and supply chain management, internet of things, and multicriteria decision-making. Various metaheuristic optimization tools are being used in these spheres of life. In this book, after a brief description of the metaheuristic tools in the first chapter, eleven more chapters have been dedicated for dealing with such applications of the optimization techniques in industrially relevant fields. The authors from different countries have shared their experience in the optimization of systems like banking, steel making, manufacturing processes, electrical vehicles design, and civil constructions. The editors express their gratitude to all the authors for their effort to make excellent contributions for the book. The editors are also grateful to all the reviewers who have taken the pain to improve the quality of the chapters further and make the book even better. The colleagues, friends, and family members of both the editors are gratefully acknowledged. The editors also acknowledge the brilliance of the Springer team shaping the compilation beautifully and express thanks to them.