Genetic Algorithm for Solving and Optimizing Assembly Sequence - Case Study Example

Genetic Algorithm for solving and optimizing assembly sequence First Author name Abstract The aim of the paper was to design genetic algorithm that was to optimize the assembly sequence problem in both large and small scale with constraints. The design of assembly sequence problem showed how encoding takes place in all industries when constraints are involved. The model accommodates all types of optimization problems facing various industries, various issues in modeling have been highlighted in order to show how genetic algorithm works as well as how it is used in real life situations. In order to achieve successive results the paper has created a classical structure which is modified to incorporate genetic operators for easier manipulation. The model in this case will only be able to solve a feasible assembly sequence problem when the parameters are provided. In the results section a case study has been used to demonstrate how it works. Key Words: Assembly sequence, genetic algorithm, Encoding 1.0 Introduction Assembly sequence is a process which all companies which are involved in the manufacture of multi-components items used. Assembly sequence is critical in this industry as it affects the cost of productivity as well as the time taken to produce the products (Mukred, Muslim and Selamat, 2013). If assembly sequence is not well managed the cost of productivity will be high as well as the quality of the products will be affected the resources that are used in an organization may not be efficiently used in production if proper modeling is not done during the automation of the production(Bai, Chen, Bin and Hun, 2005). Therefore genetic algorithm will be used to design a model that will optimize assembly sequence of this type of production leading to efficient use of resources, reduction of cost production and exploitation of time of production (Ou and Xu, 2013). The algorithm will be designed in a manner that it is able to assist production without affecting the systems and products quality. This is because it will participate in solving a problem in order to optimize the production process. The optimization algorithm accommodates assembly sequences by ensuring there is optimal solution (Ramteke and Zanwar, 2014). Genetic algorithm is critical as it is able to incorporate minor variations in design of various products (De Fazio and Whitney, 1987). It is necessary to design new genetic algorithms as the traditional genetic algorithms have less chance of providing a feasible solution (Yu and Wang, 2013). 1.1 Aims and Objectives The main aim of this research paper is to generate a genetic algorithm that will be used in solving and optimizing assembly sequence problems for a multi-products production in real life situations. The research objectives will be To review the existing literature on genetic algorithm for optimization of multiproduct assembly sequence with the intention of providing a feasible solution that can be used in real life situations. To scrutinize the interactions between assembly sequence and resource consequence in a multiproduct situation Is to design a genetic algorithm that is relevant in solving assembly sequence problem of a multi-product line. 1.2 Organizational overview This paper has been prepared with five sections which includes introduction, literature review, analysis, discussion and conclusion. The paper begins with an abstract which is the summary of the entire work that has been done, then an introduction follows highlighting the motivation for writing the paper as well ass the aims and objectives of the paper. The literature review section examines the existing literature ion the topic from various sources. Analysis provides the methods that are used in carrying out the results. The results contain a genetic algorithm with description that will be used in the industry. 2.0 Literature review 2.1 Introduction Marian et al. (2003) developed a genetic algorithm (GA) that optimised an assembly sequence planning problem (ASPP) that was exceptionally large-scale, diverse combinatorial and highly constrained. In the same way, the Genetic Algorithm is designed to cover all types of assembly components and plans. Young-Keun et al. (2009) have focused their discussion on assembly sequence planning that uses multiple criteria, which is a known time-consuming and large scale combinatorial problem. Even though the assembly sequence planning problem is already solved using various techniques of optimisation, the solution is not reliably applicable to large-scale problems. Genetic algorithm remains the most popular evolutionary computation method because it includes biological concepts in its analytical studies of systems. They have proposed the use of the multi-criteria ASP based on GA during optimisation. In addition, a precedence matrix has been proposed with a view to determining the feasible assembly sequences that are capable of satisfying the constraints of precedence. Further, a numerical example has been shown in a bid to demonstrate how the proposed algorithm performs. The experiment’s comparison results indicate that the algorithm can be used to efficiently solve the ASPP. The results also show that the algorithm can be used in any kind of ASP with multi-objective functions and large component numbers. Azman et al. (2010) have investigated how GA can be applied in getting optimum product assembly sequences. The aim was to get the minimum time that the parts take to be assembled to form a given product. It shows that the use of a single objective in optimising the assembly sequence, thereby showing the minimum time taken, is possible. Experimental results show that GA is capable of being applied in assembly sequence planning, which can be used during the design process to get faster results as compared to traditional methods. Mathew and Rao et al. (2014) came up with the Genetic Algorithm for selecting the optimal Assembly Sequence Planning problem because the same are capable of handling large search spaces, are flexible in defining constraints and they derive them using a fitness function. An approach of using a penalty function is utilized in getting the fitness value that can be used in assembly sequences. The evaluation of the penalty function is rather straightforward and simple, which is a desirable population feature–based search. 2.2 Theoretical Considerations In designing the model for genetic algorithm quality is critical in determining information processing and generation of a product in question it is determine the size of the problem to be solved by the model (Yasin, Puteh , Daud , Omar and Abdullah, 2010). In the modeling for genetic algorithm that is to solve and optimize the assembly sequence a number of issues is to be taken into consideration among them encoding which is modeling of assembly sequence, encoding and storing, making a process and encoding feedback (Chang, Tseng and Meng, 2009). The mentioned considerations are to capture needed parameters for the model to function and produce the desired results. In the modeling there is need to consider expansion of the modeling future to accommodate future changes (Hong and Cho, 1999). 2.2. 1 Encoding assembly sequence problem The main aim of encoding assembly sequence is to make sure that there is a framework that will be able to solve the problem. When encoding, various components are incorporated such a gene and term sequences (Bonneville, Perrard and Henrioud, 1995). For example in encoding genes for chromosomes the following will be adapted. 2.2. 2 The population for genetic algorithm solving and optimization The population that will be used for modeling will be identified and examined. Then questions to be answered by the model will be designed in order to simplify the complexity of the problem to be solved. The initial population will be divided according to constraints which will affect optimization. The following is the modeling of initial population diagram. The diagram above shows how modeling is done for a product in order to produce results that are desirable after operations. In the diagram, a graph of liaisons is made whereby; there is a connection between constraints and table of liaisons, assembly table and Boolean relations. From the graph, it can be noted that It is important to note that each singular term used to denote the different complex notations such as the terms point, sample, value and signal constitutes both a real part and an imaginary part. 3.0 Research methodology and Modeling The research study focused on developing a suitable optimisation algorithm for solving and optimising assembly sequence of multi-product production. Understanding of the concept of assembly sequence planning, its problems and the genetic algorithm approach (through a literature review). Provision and evaluation of the preferred solutions (such as the development of a genetic algorithm). Optimisation and validation (programming stage). Through the relevant literature studies, identification of assembly sequence planning-related issues and other important factors were identified. These factors considered complexity of product assembly processes, time (efficiency) of product assembly and mechanical instability through assembly due to weight of a product (Smith and Smith, 2002; Tseng, 2006). 4.0 Analysis and Explanation When applying genetic algorithm for the optimization that are application in symbolic expression, evolution of the initial data into many generations which requires more time to compute the optimum solution. In the calculation to save time simplification of process is done by employing the e that are simple equation and obtain the optimum coefficients for model. The level of complexity of the analysis was simplified by the use of simple equations which simplified the mathematical equation as well as providing optimum coefficient which was applied in the analysis. In modeling genetic operators are important as they help in manipulation of the problem to produce a peaceable solution. Genetic operators will have intrinsic precedence relations which process specific parameters and provide connectivity to graph of liaisons. The figure belowshows how assembly process is done for genetic operators, each component of the graph indicate how assembly of sequences is done. The analysis provides estimated results of expected functions, so additional study of function consequences between actual phenotypes should be carried out. The level of complexity of the analysis was simplified which simplified the mathematical equation as well as providing optimum coefficient which was applied in the analysis. The same text was analyzed using genetic programming to obtain optimum values for regression coefficient 5.0 Conclusions and recommendation In summary, genetic algorithm for solving and optimizing assembly sequence problem. Therefore, based on a better understanding, this report on the assembly sequence, assembly sequence planning problems, and the GA approach has been presented. The modelled genetic algorithm to solve and optimise some of these setbacks and optimisation of the solution was conducted. Solving feasible problems using the GA leads to a tremendous operational breakthrough for the multi-product companies. 6.0 References A. Yasin, N. Puteh , R. Daud , M. Omar , S.L.S. Abdullah, Product Assembly Sequence Optimisation Based on Genetic algorithm, (IJCSE) International Journal on Computer Science and Engineering, Vol. 02, No. 09, (2010), 3065-3070. A. T. Mathew, C.S.P. Rao, Implementation of genetic algorithm to optimise the assembly sequence plan based on penalty function, ARPN Journal of Engineering and Applied Sciences, Vol. 9, No. 4, (2014), 453-456. B. Lazzerini, F. Marcelloni, A genetic algorithm for generating optimal assembly plans, Artificial Intelligence in Engineering, 14, (2000), 319–329. C. Zeng, T. Gu, L. Chang, F. Li, A Novel Multi-agent Evolutionary Algorithm for Assembly Sequence Planning, Journal of Software, Vol. 8, No. 6, (2013), 1518-1525. C.C. Chang, H.E. Tseng, L.P. Meng, Artificial immune systems for assembly sequence planning exploration, Engineering Applications of Artificial Intelligence, 22, (2009), 1218–1232. D.S. Hong, H.S. Cho, A genetic-algorithm based approach to the generation of robotic assembly sequence, Control Engineering Practice, 7, (1999), 151-159. F. Bonneville, C. Perrard, J.M. Henrioud, A genetic algorithm to generate and evaluate assembly Plans, Proceedings of the IEEE Symposium on Emerging Technology and Factory Automation, Paris, (1995), 231-239. F.Y. Huang, Y.J. Tseng, An Integrated Design Evaluation and Assembly Sequence Planning Model using a Particle Swarm Optimisation Approach, World Academy of Science, Engineering and Technology, Vol. 5, (2011), 236-241. G.C. Smith, S.S.F. Smith, An enhanced genetic algorithm for automated assembly planning, Robotics and Computer Integrated Manufacturing, 18, (2002), 355–364. H. Wang, Y. Rong, D. Xiang, Mechanical assembly planning using ant colony optimisation, Computer-Aided Design, 47, (2014), 59–71. H.E. Tseng, Guided genetic algorithms for solving a larger constraint assembly problem, International Journal of Production Research, Vol. 44, No. 3, (2006), 601–625. J. Yu, C. Wang, A max–min ant colony system for assembly sequence planning, Int J Adv Manuf Technol, 67, (2013), 2819–2835. J.A.A. Mukred, M.T. Muslim, H. Selamat, Optimizing Assembly Sequence Time Using Particle Swarm Optimisation (PSO), Applied Mechanics and Materials, Vol. 315, (2013), 88-92. J.D. Wolter, A Constraint Based Approach to Planning with Subassemblies. Planning, Proceedings of the IEEE International Conference on Systems Engineering, (1990), 412-415. J.H. Holland, Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence. Ann Arbor, MI: University of Michigan Press, (1975), 1-183. J.H. Holland, Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence, 2nd ed. Cambridge, MA: MIT Press, (1992), 1-211. L. shuxia, S. Hongbo, GSSA and ACO for Assembly Sequence Planning: A Comparative Study, Proceedings of the IEEE International Conference on Automation and Logistics Qingdao, China September, (2008), 1270-1275. L. Wang, S. Keshavarzmanesh, H.Y. Feng, R.O. Buchal, Assembly process planning and its future in collaborative manufacturing: a review, Int J Adv Manuf Technol, 41, (2009), 132–144. L. Wang, Y. Hou, X. Li, S. Sun, An enhanced harmony search algorithm for assembly sequence planning, Int. J. Modelling, Identification and Control, Vol. 18, No. 1, (2013), 18-25. L.M. Ou, X. Xu, Relationship matrix based automatic assembly sequence generation from a CAD model, Computer-Aided Design, 45, (2013), 1053–1067. M. F. Sebaaly, H. Fujimoto, Agenetic planner for assembly automation, Proceedings of the IEEE International Conference on Evolutionary Computation, (1996), 401–406. N. Senin, R. Groppetti, D.R. Wallace, Concurrent assembly planning with genetic algorithms, Robotics and Computer Integrated Manufacturing, 16, (2000), 65-72. P.E. Ramteke, D.R. Zanwar, Assembly Sequence Optimisation, International Journal of Innovative Research in Advanced Engineering (IJIRAE), Volume 1, Issue 6, (2014), 15-22. R. M. Marian, L.H.S. Luong, K. Abhary, A genetic algorithm for the optimisation of assembly sequences, Computers & Industrial Engineering, 50, (2006), 503–527. R.M. Marian, L.H.S. Luong, K. Abhary, Assembly sequence planning and optimisation using genetic algorithms Part I. Automatic generation of feasible assembly sequences, Applied Soft Computing, 2/3F, (2003), 223–253. R. E. Jones, R.H. Wilson, T.L. Calton, Constraint-Based Interactive Assembly Planning, Proceedings of the IEEE International Conference on Robotics and Automation Albuquerque, New Mexico, April (1997), 913-920. S. Hongbo, L. Shuxia, G. Degang, L. Peng, Genetic Simulated Annealing Algorithm-Based Assembly Sequence Planning, International Technology and Innovation Conference, (2006), 1573-1579. S. Hongbo, L. shuxia, The Comparison Between Genetic Simulated Annealing Algorithm and Ant Colony Optimisation Algorithm for ASP, Proceedings of the IEEE International Conference on Automation and Logistics, (2008), 1-6. S. Zhao, J. Hong, H. Zhao, S. Li, and L. Ding, Research on assembly sequence planning method for large-scale assembly based on an integrated assembly model, Proceedings of the Institution of Mechanical Engineers, Part B: Journal of Engineering Manufacture, Vol. 226, No. 4, (2012), 733-744. T.L.D. Fazio, D.E. Whitney, Simplified Generation of All Mechanical Assembly Sequences, IEEE Journal of Robotics and Automation, Vol. RA-3, No. 6, (1987), 640-658. W. Zhou, J. Yan, Y. Li, C. Xia, J. Zheng, Imperialist competitive algorithm for assembly sequence planning, Int J Adv Manuf Technol, 67, (2013), 2207–2216. W.P. Wang, Complexity Estimation for Genetic Assembly Sequence Planning, Journal of the Chinese Institute of Industrial Engineers, Vol. 26, No. 1, (2009), 44-52. Y. Xing, Y. Wang, Assembly sequence planning based on a hybrid particle swarm optimisation and genetic algorithm, International Journal of Production Research, Vol. 50, No. 24, (2012), 7303–7312. Y.J. Tseng, H.T. Kao and F.Y. Huang, Integrated assembly and disassembly sequence planning using a GA approach, International Journal of Production Research, Vol. 48, No. 20, (2010a), 5991–6013. Y.J. Tseng, J.Y. Chen, F.Y Huang, A multi-plant assembly sequence planning model with integrated assembly sequence planning and plant assignment using GA, Int J Adv Manuf Technol, 48, (2010b), 333–345. Y.W. Bai, Z.N. Chen, H.Z. Bin, J. Hun, An effective integration approach toward assembly sequence planning and evaluation, Int J Adv Manuf Technol, 27, (2005), 96–105. Z. Xiaoming, D. Pingan, A model-based approach to assembly sequence planning, Int J Adv Manuf Technol, 39, (2008), 983–994. Q Su Applying case-based reasoning in assembly sequence planning. Int J Prod Res 45(2007):29–47 R.M. Marian, L.H.S. Luong & Abhary, K. A genetic algorithm for the optimisation of assembly sequences. Computers & Industrial Engineering 50 (2006) 503–527 Read More

The experiment’s comparison results indicate that the algorithm can be used to efficiently solve the ASPP. The results also show that the algorithm can be used in any kind of ASP with multi-objective functions and large component numbers. Azman et al. (2010) have investigated how GA can be applied in getting optimum product assembly sequences. The aim was to get the minimum time that the parts take to be assembled to form a given product. It shows that the use of a single objective in optimising the assembly sequence, thereby showing the minimum time taken, is possible.

Experimental results show that GA is capable of being applied in assembly sequence planning, which can be used during the design process to get faster results as compared to traditional methods. Mathew and Rao et al. (2014) came up with the Genetic Algorithm for selecting the optimal Assembly Sequence Planning problem because the same are capable of handling large search spaces, are flexible in defining constraints and they derive them using a fitness function. An approach of using a penalty function is utilized in getting the fitness value that can be used in assembly sequences.

The evaluation of the penalty function is rather straightforward and simple, which is a desirable population feature–based search. 2.2 Theoretical Considerations In designing the model for genetic algorithm quality is critical in determining information processing and generation of a product in question it is determine the size of the problem to be solved by the model (Yasin, Puteh , Daud , Omar and Abdullah, 2010). In the modeling for genetic algorithm that is to solve and optimize the assembly sequence a number of issues is to be taken into consideration among them encoding which is modeling of assembly sequence, encoding and storing, making a process and encoding feedback (Chang, Tseng and Meng, 2009).

The mentioned considerations are to capture needed parameters for the model to function and produce the desired results. In the modeling there is need to consider expansion of the modeling future to accommodate future changes (Hong and Cho, 1999). 2.2. 1 Encoding assembly sequence problem The main aim of encoding assembly sequence is to make sure that there is a framework that will be able to solve the problem. When encoding, various components are incorporated such a gene and term sequences (Bonneville, Perrard and Henrioud, 1995).

For example in encoding genes for chromosomes the following will be adapted. 2.2. 2 The population for genetic algorithm solving and optimization The population that will be used for modeling will be identified and examined. Then questions to be answered by the model will be designed in order to simplify the complexity of the problem to be solved. The initial population will be divided according to constraints which will affect optimization. The following is the modeling of initial population diagram.

The diagram above shows how modeling is done for a product in order to produce results that are desirable after operations. In the diagram, a graph of liaisons is made whereby; there is a connection between constraints and table of liaisons, assembly table and Boolean relations. From the graph, it can be noted that It is important to note that each singular term used to denote the different complex notations such as the terms point, sample, value and signal constitutes both a real part and an imaginary part. 3.0 Research methodology and Modeling The research study focused on developing a suitable optimisation algorithm for solving and optimising assembly sequence of multi-product production.

Understanding of the concept of assembly sequence planning, its problems and the genetic algorithm approach (through a literature review). Provision and evaluation of the preferred solutions (such as the development of a genetic algorithm). Optimisation and validation (programming stage). Through the relevant literature studies, identification of assembly sequence planning-related issues and other important factors were identified.

Read More