The auto body process monitoring and the root cause diagnosis based on data-driven approaches are
vital ways to improve the dimension quality of sheet metal assemblies. However, during the launch time of the
process mass production with an off-line measurement strategy, the traditional statistical methods are difficult to
perform process control effectively. Based on the powerful abilities in information fusion, a systematic Bayesian
based quality control approach is presented to solve the quality problems in condition of incomplete dataset. For
the process monitoring, a Bayesian estimation method is used to give out-of-control signals in the process. With
the abnormal evidence, the Bayesian network (BN) approach is employed to identify the fixture root causes. A
novel BN structure and the conditional probability training methods based on process knowledge representation
are proposed to obtain the diagnostic model. Furthermore, based on the diagnostic performance analysis, a case
study is used to evaluate the effectiveness of the proposed approach. Results show that the Bayesian based method
has a better diagnostic performance for multi-fault cases.
LIU Yinhua1* (刘银华), YE Xialiang1 (叶夏亮), JIN Sun2 (金隼)
. A Bayesian Based Process Monitoring and Fixture Fault Diagnosis Approach in the Auto Body Assembly Process[J]. Journal of Shanghai Jiaotong University(Science), 2016
, 21(2)
: 164
-172
.
DOI: 10.1007/s12204-016-1708-1
[1] CEGLAREK D, SHI J. Fixture failure diagnosis for autobody assembly using pattern recognition [J]. Journal of Engineering for Industry, 1996, 118(1): 55-66.
[2] LIU Y G, HU S J. Assembly fixture fault diagnosis using designated component analysis [J]. Journal of Manufacturing Science and Engineering, 2005, 127(2):358-368.
[3] DU S, L¨U J, XI L. A robust approach for root causes identification in machining processes using hybrid learning algorithm and engineering knowledge[J]. Journal of Intelligent Manufacturing, 2012, 23(5):1833-1841.
[4] APLEY D W, SHI J. Diagnosis of multiple fixture faults in panel assembly [J]. Journal of Manufacturing Science and Engineering, 1998, 120(4): 793-801.
[5] LIU S C, HU S J. Variation simulation for deformable sheet metal assemblies using finite element methods[J]. Journal of Manufacturing Science and Engineering,1997, 119(3): 368-374.
[6] CAMELIO J A, HU S J, YIM H. Sensor placement for effective diagnosis of multiple faults in fixturing of compliant parts [J]. Journal of Manufacturing Science and Engineering, 2005, 127(1): 68-74.
[7] YU K G, JIN S, LAI X M. Fixture variation diagnosis of compliant assembly using sensitivity matrix[J]. Journal of Shanghai Jiaotong University (Science),2009, 14(6): 707-712.
[8] DING Y, CEGLAREK D, SHI J. Fault diagnosis of multistage manufacturing processes by using state space approach [J]. Journal of Manufacturing Science and Engineering, 2002, 124(2): 313-322.
[9] ZHOU S, CHEN Y, SHI J. Statistical estimation and testing for variation root-cause determination of multistage manufacturing processes [J]. IEEE Transactions on Automation Science and Engineering, 2004, 1(1):73-83.
[10] DEY S, SRORI J A. A Bayesian network approach to root cause diagnosis of process variations [J]. International Journal of Machine Tools & Manufacture, 2005,45(1): 75-91.
[11] JIN S, LIU Y, LIN Z. A Bayesian network approach for fixture fault diagnosis in launch of the assembly process[J]. International Journal of Production Research,2012, 50(23): 6655-6666.
[12] LAURITZEN S L. The EM algorithm for graphical association models with missing data [J]. Computational Statistics & Data Analysis, 1995, 19(2): 191-201.
