Advances in Production Engineering & Management
Volume 12 | Number 3 | September 2017 | pp 285–295
https://doi.org/10.14743/apem2017.3.259
Predicting the availability of production lines by combining simulation and surrogate model
Jia, Y.; Tian, H.; Chen, C.; Wang, L.
ABSTRACT AND REFERENCES (PDF) |
FULL ARTICLE TEXT (PDF)
A B S T R A C T
The availability analysis plays a significant part in both the design and operations management of production lines. In this paper, a method combining discrete event simulation (DES) and surrogate model is presented to predict the availability of production lines with unreliable workstations and finite intermediate buffers. The DES can conduct computer experiments for production lines with the help of design of experiments (DOE) under the Matlab environment. The surrogate model is constructed by using Kriging model integrated with Latin hypercube sampling (LHS), which can predict the responses based on a limited set of simulation results. The major advantages of the proposed approach are its flexibility and convenience. Also, it is the first time to investigate Kriging opportunities in predicting the performance of production lines. Finally, an application in a crankshaft production line is presented, and the results indicate that the proposed approach can achieve higher prediction accuracy than the other methods.
A R T I C L E I N F O
Keywords • Production lines, Availability prediction, Discrete event simulation (DES), Kriging model, Latin hypercube sampling (LHS)
Corresponding author • Chen, C.
Article history • Received 5 May 2017, Revised 22 June 2017, Accepted 10 July 2017
Published on-line • 28 July 2017
E X P O R T C I T A T I O N
» RIS format (EndNote, ProCite, RefWorks, and most other reference management software)
» BibTeX (JabRef, BibDesk, and other BibTeX-specific software)
» Plain text
