Generating Large Data Sets for Simulation of Electronics Manufacturing

Ping Zhang, James B. Pick

Research output: Contribution to journalArticle

1 Scopus citations

Abstract

Very often the data sets needed for large-scale system simulation and testing aren't available. Even when it's possible to collect and use the real-world data, they're not always suitable. In some situations, only a small portion of the data sets is actually needed for system testing. In others, the sets may involve many data variables and extensive data elements in each data variable, creating high complexity and difficulty. This is especially true in manufacturing production planning, where many factors must be considered, and the scope of the data sets is often very large. Here we introduce the procedure and methods we developed for generating large data sets in manufacturing using Monte Carlo techniques combined with the Extended Entity Relationship modeling method. We introduce an approach that can deal with complicated relationships and ordering among random variates. We generate the data sets for an IBM electronics manufacturing facility. We examine use of the sets to test an information visualization system for production planning. We discuss the goals of random sample generation and the verification of the generation of the random variates.

Original languageEnglish (US)
Pages (from-to)231-249
Number of pages19
JournalSimulation
Volume70
Issue number4
StatePublished - 1998

    Fingerprint

Keywords

  • Data modeling
  • Manufacturing
  • Monte Carlo
  • Verification

ASJC Scopus subject areas

  • Computational Theory and Mathematics
  • Computer Graphics and Computer-Aided Design
  • Computer Science Applications
  • Software
  • Safety, Risk, Reliability and Quality

Cite this