Big data analysis and scheduling optimization system oriented assembly process for complex equipment

ABSTRACT:

Traditionally, the data generated in the manufacturing process is not under full use for management decisions, so it is difficult to achieve the decision optimization in a manufacturing system level. The analysis of the big data with uncertain information that influences the assembly objectives can be helpful to improve the capacity of resisting the disturbance of the scheduling system and realize the optimal production. This paper focuses on the analysis and utilization of assembly big data in manufacturing process, and studies the key technologies of big data analysis and assembly scheduling optimization for complex equipment. A big data analysis and scheduling optimization system is proposed to solve the assembly service execution decision for complex equipment. It proposes to analyze the assembly big data and make decisions with uncertain information by Locally Linear Embedding, Adaptive Boosting, Support Vector Machine and D-S evidence theory et al. In order to explore the influence pattern of assembly task and environment on assembly efficiency, the neural industrial engineering is proposed to be introduced into human error prediction based on physiological big data. Finally, the variable metric clustering of assembly tasks can be provided to ensure the maximization of assembly efficiency and the production balance. The proposed system can effectively handle the dynamic and uncertain information in the assembly, and get better overall scheduling optimization of the assembly system. The support technologies presented in this paper can provide a good theoretical foundation and application reference for big data decision to be used in manufacturing optimization.

Existing System:

The heterogeneous data generated during assembly process makes the assembly scheduling optimization problem more complicated. Many common methods and models have limitations on dealing with assembly big data. The assembly scheduling based on big data is essentially a multi-objective NP problem, which refers to multi-objective optimization with constraints of assembly resource capabilities. However, influence of diverse factors is hard to be described by a clear mathematical model. In fact, assembly scheduling involves a series of problems, such as assembly data acquisition and processing, assembly quality factor analysis, quality evaluation, multi-objective job scheduling optimization, line balancing, etc. On the other hand, the development of data mining makes the optimal scheduling based on assembly big data feasible.

Proposed System:

The traditional scheduling model based on basic manufacturing data, the big data analysis and scheduling optimization system presented in this paper has the advantage of big data mining. It includes the dynamic and uncertain information in the assembly execution, which helps to get a more accurate and comprehensive assembly evaluation and optimization result. This paper also provides a way to solve the Curse of Dimensionality. In some cases, the small-batch production of complex equipment may lead to the shortage of historical manufacturing data and affects the optimization effect. In the future, it is required to improve the model and algorithm to better adapt to insufficient data in the proposed system.

Main Goal:

  1. a) to establish a decision table after invalid or weak efficient data reduction; b) to complete the incomplete information; c) consistency processing for conflicting and inconsistent information.

Conclusion:

A multi-objective decision making and scheduling optimization platform is proposed for complex equipment assembly. The paper also expounds the support technologies of the big data analysis and scheduling optimization system to provide decision-making support for the quantitative and information-based equipment manufacturing.

REFERENCES:

[1] H. L. Zhao, “The study of CNC machine tool’s assembly reliability based on assembly complexity,” M.S. thesis, Chongqing University, Chongqing, China, 2014.

[2] M. Z. Liu, Z. B. Zhao, M. G. Ge, X. Q. Wang, and L. Wang, “Online optimization for assembly quality of complex mechanical products based on improved PSO,” Transactions of the Chinese Society for Agricultural Machinery, vol. 44, no. 6, pp. 246-252, 2013. [Online]. Available: http://dx.doi.org/10.6041/j.issn.1000- 1298.2013.06.043.

[3] Q. Wang, L. Q. Wen, J. X. Li, Y. L. Ke, T. Li, and S. J. Zhang, “Modeling and optimization for aircraft final assembly line based on Petri net,” Journal of Zhejiang University (Engineering Science), vol. 49, no. 6, pp. 1224-1231, 2015. DOI: 10.3785/j.issn.1008-973X.2015.07.004.

[4] Y. H. Liu, D. W. Zuo, and D. Zhang, “Assembly process modeling method based on matrix technique,” Journal of Nanjing University of Aeronautics & Astronautics, vol. 46, no. 6, pp. 938-944, 2014. [Online]. Available: http://dx.doi.org/10.3969/j.issn.1005- 2615.2014.06.018.

[5] B. Wang, X. Q. Tang, and R. J. Geng, “Modeling for assembly structure in mechanical product,” Journal of Beijing University Aeronautics and Astronautics, vol. 36, no. 1, pp. 71-76, 2010. [Online]. Available: http://dx.doi.org/10.13700/j.bh.1001- 5965.2010.01.005.

[6] E. G. Minca, V. Stefan, A. Filipescu, and A. Serbencu, “Two approaches in modeling of assembly/disassembly line with integrated manipulator mounted on mobile platform,” IEEE ICSTCC, Sinaia, Romania, pp. 1-6, 2012. [Online]. Available: http://ieeexplore.ieee.org/xpls/abs_all.jsp?arnumber=6379285.

[7] T. Aized, K. Takahashi, and I. Hagiwara, “Modeling and optimization of multiple product FMS using colored Petri net and design of experiment,” IEEE International Conference on Networking, Sensing and Control, pp. 334-339, 2007. DOI: 10.1109/ICNSC.2007.372801.

[8] J. Miltenburg, “Balancing and scheduling mixed-model u-shaped production lines,” International Journal of Flexible Manufacturing Systems, vol. 14, no. 2, pp. 119-151, 2002. [Online]. Available: http://dx.doi.org/10.1023/A:1014434117888.

[9] J. Pereira and M. Vilà, “An exact algorithm for the mixed-model level scheduling problem,” International Journal of Production Research, vol. 53, no. 19, pp. 5809-5825, 2015. [Online]. Available: https://doi.org/10.1080/00207543.2015.1005771.

[10] G. M. Jiao, D.W. Zuo, D. Zhang, and M. Wang, “Scheduling models of aerospace product agile mixed-model assembly,” Journal of Southeast University (Natural Science Edition), vol. 40, no. 4, pp. 717-721, 2010. [Online]. Available: http://kns.cnki.net/KCMS/detail/detail.aspx?filename=dndx2010040 11&dbname=CJFD&dbcode=CJFQ.

[11] R. K. Hwang and H. Katayama, “Sequencing problem for mixedmodel assembly line in a manual boiler plant,” IEEE CIE, July, 2010, pp. 1-5. [Online]. Available: http://dx.doi.org/10.1109/ICCIE.2010.5668430.

[12] J. Bautista and A. Cano, “Solving mixed model sequencing problem in assembly lines with serial workstations with work overload minimisation and interruption rules,” European Journal of Operational Research, vol. 210, no. 3, pp. 495-513, 2011. [Online]. Available: http://dx.doi.org/10.1016/j.ejor.2010.10.022.

[13] X. Z. Pi, X. M. Fan, and J. Q. Yan, “Application of genetic algorithm based on task sequences to assembly line balancing,” Mechanical Science and Technology, vol. 22, no. 1, pp. 35-38, 2003. [Online]. Available: http://dx.doi.org/10.3321/j.issn:1003- 8728.2003.01.012.