In this research, an analytical model for analyzing a production line consisting of a series
of automated workstations with infinite buffers is developed. Automated workstations are
assumed to have deterministic processing times, and independent exponentially
distributed operating time between failures and repair times. The analytical model starts
with existing results from a Markov chain model of two automated workstations in series,
where analytical expressions are developed for the average number of jobs in the second
workstation and its queue. This research focuses on the development of a set of linking
equations that can be used to analyze larger systems using a two workstation
decomposition approach. These linking equations utilize probabilities computed in the
two-workstation Markov chain model to compute workstation parameters for a single
workstation such that the first two moments of the inter-departure process from the two-workstation
system and the single workstation are the same. Simulations of a number of
different 3-workstation and 10-workstation systems were carried out employing a range
of workstation utilizations and processing time coefficients of variation. The results from
these simulations were compared with those calculated with the analytical model and
various two-parameter GI/G/1 approximations and linking equations present in the
literature. The analytical model resulted in an average absolute percentage difference of
less than 5% in the systems studied, and performed much better than general two parameter
G/G/1 approximations. The analytical model was also robust in ranking the
queues in the order of the average number of jobs present in the queues. / Graduation date: 2004
Identifer | oai:union.ndltd.org:ORGSU/oai:ir.library.oregonstate.edu:1957/32257 |
Date | 08 December 2003 |
Creators | Nagarajan, Raghavendran D. |
Contributors | Kim, David S. |
Source Sets | Oregon State University |
Language | en_US |
Detected Language | English |
Type | Thesis/Dissertation |
Page generated in 0.0081 seconds