e-ISSN: 2582-5208 International Research Journal of Modernization in Engineering Technology and Science Volume:03/Issue:06/June-2021 Impact Factor- 5.354 www.irjmets.com www.irjmets.com @International Research Journal of Modernization in Engineering, Technology and Science [1241] APPLICATION OF GENETIC ALGORITHM FOR SCHEDULING PROBLEM IN MANUFACTURING SECTOR Deepanshu Arora *1 , Sunil Kumar Jakhar *2 , Sanjay Choudhary *3 , Krishna Nandan *4 , Amogh Sharma *5 *1,2,3,4,5 Assistant Professor, Department Of Mechanical Engineering, Vivekananda Institute Of Technology, Jaipur, Rajasthan, India. ABSTRACT The present work is based on the various types of evolutionary approaches used for solving the flow shop scheduling problem (flow shop scheduling problem with sequence dependent setup time). Here we focused on evolutionary approach Genetic Algorithm as a solution technique. Study is based on a Genetic Algorithm (GA) approach for Flow Shop Scheduling (FSP) using a direct, time-based representation. Since Flow Shop Scheduling is a multi-objective optimization problem so the Genetic Algorithm (GA) is an evolutionary approach for performing optimizing FSP. Whereas traditional solution methods are typically sequence based, this representation encodes schedule information as mathematical form. The GA approach is Schedule permutation approach, in which a population of Schedules is initially generated, and through some operations, good traits of these Schedules are used for producing better schedules. Keywords: Sequence Dependent Setup Time (SDST), Flow Shop Scheduling Problem (FSSP), Meta-Heuristics, Genetic Algorithm (GA), Tardiness, Earliness, Makespan, Tardy Jobs. I. INTRODUCTION The Flow Shop Scheduling and genetic algorithm (on macro level) both have dynamic property. As we know is a system which is unstable, and it carries variables which are dependent on some feature or parameters of environment. In Flow Shop Scheduling the scheduling of job is unstable; we cannot say what will be the next place for job after completing some operation for better result, because it depends upon various parameter of system of Flow Shop like cycle time, lead time, arrival of time of job etc. Same as at the macro level of Genetic Algorithm have same property; during the creation of new generation (parent to child) we cannot say what characters (like color of eyes, type of hair, etc.) of past or parent generation will be carried by the child or next generation. By this way they both have same dynamic property. With the knowledge and information of genetic coding of parent. characters we can define what characters will be carried by next generation and can understand the creation of new generation very easily. That is the main logic which is follow by Genetic Algorithm for solving real time problems like Flow Shop Scheduling. By this we convert the decision variable of Flow Shop scheduling in for of genetic coding and following the operations of genetic like reproduction, crossover, and mutation and we can solve find more optimized results for Flow shop Scheduling problem in very easy way. Single and Multi-Objective Optimization: When an optimization problem modeling a physical system involves only one objective function, the task of finding the optimal solution is called “Single Objective Optimization”. The optimization problem involves more than one objective function, the task of finding one or more optimum solutions is known as “Multi Objective Optimization”. Fundamental Difference between Single and Multi - Objective Optimization is that in “Single Objective Optimization” has a single solution but in “Multi Objective Optimization” have sets of solutions Evolutionary Algorithm for Multi Objective Optimization: Evolutionary Algorithms (EAs) have been used widely to solve multi-objective optimization problems. The optimization can be divided into two types continuous and combinational. Further Continuous optimization can be divided into three types are linear, quadratic and nonlinear, whereas combinational can be divided into two types of i.e. approximate method and exact method. Next to this the approximate method has two types of methodologies are heuristics and meta-heuristics for the solution of the problem. The beauty of the meta- heuristics is that it can handle Nonlinear problem very easily. If we talk about the latest scenario search methods, the meta heuristics are the combinations of the randomization with the traditional heuristic methods.