Sensor Networks for Smart Manufacturing Processes

Abstract:

Each factory and manufacturing plant needs a flexible and reliable in-plant resource supply to serve production processes efficiently. Manufacturing systems are composed of several numbers of elements, workstations, machines and logistics resources. Production line is a complex system because of the complicated manufacturing process, multiple types, high machining difficulty and many special processes in it. In the Industry 4.0 based on smart manufacturing, it is essential to support the processes with intelligent sensor networks. In this article, we give a brief overview about sensors often used in manufacturing processes. Sensor networks generate a massive and increasing amount of data that needs to be processed. Computationally intensive algorithms are used for the data processing (image, voice and signal processing, different classification functions, numeric optimization routines). Finally, we discuss how GPGPU can improve the real-time processing of data generated by intelligent sensor networks.

• Periodical: Solid State Phenomena (Volume 261)
• Pages: 456-462
• DOI: https://doi.org/10.4028/www.scientific.net/SSP.261.456
• Online since: August 2017
• Authors: József Konyha, Tamás Bányai*
• Keywords: GPGPU, Intelligent Sensor Networks, Parallel Processing, Smart Manufacturing

References

[1] J. Korponai, A. Banyaine Toth, B. Illes, The effect of the safety stock on the occurrence probability of the stock shortage, Manag. Prod. Eng. Rev. 8(1) (2017) 69-77.

DOI: 10.1515/mper-2017-0008

[2] E. A. Lee, Cyber physical systems: Design challenges, Proc. Obj. Oriented RT. Dist. Comp. (2008) IEEE 363–369.

[3] W. Wolf, Cyber physical systems, Computer 42(3) (2009) 88-89.

[4] J. Fraden, Handbook of Modern Sensors - Physics, Designs, and Applications, fifth ed. Switzerland, (2016).

[5] S. Mekid, Further Structural Intelligence for Sensors Cluster Technology in Manufacturing, Sensors 6(6) (2006) 557-577.

DOI: 10.3390/s6060557

[6] I. F. Akyildiz, W. Su, Y. Sankarasubramaniam, E. Cayirci, Wireless sensor networks: a survey, Comput. Networks 38(4) (2002) 393-422.

DOI: 10.1016/s1389-1286(01)00302-4

[7] T. L. Johnson, M. E. Dausch, Sensor Informatics for Manufacturing, P. IFAC Proc. Vol. 39, (2006) 125-130.

[8] Y. Hao, P. Helo, The role of wearable devices in meeting the needs of cloud manufacturing: A case study, Robot. CIM-Int. Manuf. 45 (2017) 168-179.

DOI: 10.1016/j.rcim.2015.10.001

[9] Theorin, K. Bengtsson, J. Provost, M. Lieder, T. Johnsson, T. Lundholm, B. Lennartson, An event-driven manufacturing information system architecture for Industry 4. 0, Int. J. Prod. Res. 55(5) (2017) 1297-1311.

DOI: 10.1080/00207543.2016.1201604

[10] J. Konyha, T. Banyai, Approach to accelerate algorithms to solve logistic problems with GPGPU, J. Adv. Log. Sys. 10 (2016) 5-10.

[11] V. Fuvesi, J. Konyha, Review of machine learning toolboxes, P. M. Tud. EM (2016) 116-122.

[12] F. Lopez, L. Zhang, A Mok, J Beaman, Particle filtering on GPU architectures for manufacturing applications, Comp. in Industry 71 (2015) 116–127.

DOI: 10.1016/j.compind.2015.03.013

[13] D. Wayland, R. Montemanni, L. M. Gambardella, A metaheuristic framework for stochastic combinatorial optimization problems based on GPGPU with a case study on the probabilistic traveling salesman problem with deadlines, J. Parallel Distrib. Comput. 73 (2013).

DOI: 10.1016/j.jpdc.2012.05.004

[14] I. M. Coelho, V. M. Coelho, E. J. S. Luz, L. S. Ochi, F. G. Guimarães, E. Rios, A GPU deep learning metaheuristic based model for time series forecasting, Applied Energy (2017) In P.

DOI: 10.1016/j.apenergy.2017.01.003