A new approach for data processing in supply chain network based on FPGA. Issue 1 (April 2016)

• Record Type: Journal Article
• Authors: Zou, Xiaofu; Tao, Fei; Jiang, Penglong; Gu, Shixin; Qiao, Kan; Zuo, Ying; Xu, Lida
• Is Part Of: International journal of advanced manufacturing technology. Volume 84:Issue 1/4(2016)
• Page Start: 249
• Page End: 260
• Publication Date: 2016-04
• Subjects: Supply chain network (SCN) -- Risk -- Field programmable gate array (FPGA) -- Kalman filter -- Median filter -- Real-time -- Big data; Manufacturing processes -- Periodicals; Production engineering -- Periodicals; 670.427
• DOI: 10.1007/s00170-015-7803-x
• Languages: English
• ISSNs: 0268-3768

Abstract

Abstract With the development of the supply chain network (SCN) and big data processing, a simple view at above two technologies separately has become unadvisable; this mainly reflected on the growing amount of data in the SCN nodes. Since the data processing is not timely, the potential risks may spread like dominoes to the entire network. There are many small nodes in the SCN, in which it requires data processing equipment that have characteristics of high integration and miniaturization. For the phenomenon that large shape and real-time shortage exists in designing underlying data processing equipment in SCN with the traditional special large computers and general embedded processor, this paper considering the advantage of parallel character of FPGA (field programmable gate array), a new method, which uses FPGA to design embedded device to execute related algorithms in parallel for underling data processing to reduce risks which caused by time-delay in SCN, is proposed. To verify effectiveness of the proposed method, a FPGA-based design method for Kalman filter algorithm and median filter algorithm is proposed. The performance and advantage of the proposed method are analyzed by comparing with traditional methods. The results shown that the proposed method have high real-time and accuracy in complex algorithm execution for SCN underling data processing, which can reduce the potential risk and improve the performance of entire SCN.