Abstract
Multi-objective optimization is one of the most popular research areas in the world of manufacturing. It concerns the manufacturing optimization problems involving more than one optimization simultaneously, but in this present scenario, it is becoming very tough to solve a manufacturing-related multi-objective problem as no logical method has been developed in assignment of response individual weight. Therefore, to tackle this problem, this chapter proposes a new integrated approach by combining Standard Deviation Method with Particle Swarm Optimization. Two examples of optimizing the advanced manufacturing process parameters are performed to test the proposed approach. The examples considered for this approach are also attempted using other established optimization techniques such as Desirability-based RSM and SDM-GA. The results verify the effectiveness of the proposed approach during multi-objective manufacturing process parameter optimization.
TopIntroduction
Manufacturing is the process of turning of raw material into finished product. Manufacturing of any product involves various aspects, such as designing of product, economic analysis of product or process and controlling of quality of product during production.
The designing of product based upon the functional requirement, material available which best suits the application and manufacturing process available for them. Monitoring and controlling of product quality is performed before and during operation to ensure robustness of product to customer. Further, economic analysis of manufacturing process is a very important consideration.
Manufacturing process is the fundamental subject since it is of interest not only to mechanical engineers but also to those forms practically every discipline of engineering. It is so because engineering as a whole meant for providing various material for human consumption. For various products such as plant machineries of civil, mechanical, chemical, textile industries, etc, the manufacturing process forms a vital ingredient (Rao, 2011). However the entire process has been slow and costly, requiring substantial human and material resources. In order to make the process fast, now-a-days many large industries use highly automated and computer-controlled machine which enhances the capital as well as manufacturing cost of the product. Therefore, to obtain the required payback it is necessary to operate these machines as efficiently as possible. For this purpose a number of parameters are needed to be satisfied. The success of the manufacturing process depends on the selection of appropriate process parameters. The selection of optimum process parameters play an important role to ensure quality of product, to reduce the manufacturing cost and to increase productivity in computer controlled manufacturing process.
Modeling and optimization of process parameters of any manufacturing process is generally a difficult job where the following aspects are required: knowledge of manufacturing process, empirical equations to develop realistic constrains, specification of machine capabilities, development of effective optimization criteria, and knowledge of different mathematical and numerical optimization techniques. A human process planner selects proper parameters using his own experiences of from related handbooks. However, because of the many variables, complexity and stochastic nature of the process, achieving the optimal performance, even for a highly skilled operator is rarely possible. Therefore an efficient way to solve this problem is to develop the relationship between the performance of the process and its controllable input parameters by modeling the process through suitable mathematical techniques and optimization using suitable optimization algorithm (Raj, 2012).
Optimization is the method of getting best results subjected to various resources constrains. This can be broadly classified into two categories: Conventional optimization technique and Non-conventional optimization technique. The conventional optimization algorithms are deterministic algorithms with specific rules for moving from one to another solution. The example of these algorithms includes non-linear programming, geometric programming, quadratic programming, dynamic programming etc. But, the optimization problems related to the manufacturing are generally complex in nature and characterized by mixed continuous-discrete variables and discontinuous and non-convex design spaces. Hence, conventional optimization techniques fail to give global optimum solution, as they are usually trapped in the local optima. Also the convergence of these techniques is very slow. In order to overcome these problems, researchers have proposed non-conventional techniques for optimization of process parameters of various manufacturing processes.
Key Terms in this Chapter
Multi-Objective Optimization: Also recognized as multi criteria optimization or multi attribute optimization, is an arena of multiple criteria decision making, which is concerned with mathematical optimization problems entailing more than one objective function to be optimized simultaneously.
Meta-Heuristic: Defined as an iterative generation process which steers a subordinate heuristic by combining intelligently different concepts for exploring and exploiting the search space, learning strategies are used to structure information in order to find efficiently near-optimal solutions.
Heuristic Search: A heuristic is a technique that improves the efficiency of a search process, possibly by sacrificing claims of completeness. Heuristics are like tour guides. They are good to the extent that they point in generally interesting directions; they are bad to extent that they may miss points of interest to particular individuals. Some heuristics help to guide a search process without sacrificing any claims to completeness that the process might previously have had.
Optimization: The process of maximizing desired qualities, and minimizing undesirable one.
Particle Swarm Optimization: A computational method that optimizes problem within the defined searching space by using the moment and intelligence of swarms.
Multi Performance Index: A value used to represent the aggregated response parameters.
Standard Deviation Method: A method associated with statistics and probability theory which is used to find the variation or dispersion of each variance from the average exists. It is most widely used to measure dispersion of a series and is commonly denoted by the symbol ‘s’. The standard deviation method is defined as the square-root of the average of squares of deviations, when such deviations for the values of individual items in a series are obtained from the arithmetic average. The mathematical representation of this method is as follows: or in case of frequency distribution where, X i = i th value of the variable X; = Arithmetic average; n= Number of items; f i = The frequency of the i th item.
Genetic Algorithm: Adaptive heuristic search algorithm based on the principle of natural selection and natural genetics. In order to arrive at optimal solution for design problems, the GA has been implemented so that the fundamental concepts of reproduction, chromosomal crossover, occasional mutation of genes and natural selection are reflected in the different stages of the genetic algorithm process. Although randomized, Genetic Algorithm is by no means random, instead they exploit historical information to steer the search into the region of better public presentation within the search distance. The process is initiated by selecting a number of candidate design variables either randomly or heuristically in order to create an initial population, which is then encouraged to evolve over generations to produce new designs, which are better or filter.