Autocorrelation effects in manufacturing systems performance: A simulation analysis

Abstract

Autocorrelation has been pointed out as one of the most challenging issues in manufacturing systems modeling. Numerical experimentation has shown that it may either enhance or harm performance. Furthermore, there is not yet a general agreement in what a realistic autocorrelation model is or whether it is actually relevant for practical applications. This paper provides a simulation analysis of the effects on performance caused by manufacturing process parameters following autoregressive (AR) processes. AR time series are employed for modeling variations in parameters that happen at a time scale different from the corresponding to process cycle execution. Three basic configurations are analyzed: serial line, assembly process and a disassembly process. A case study from the natural slate tiles industry is presented showing the differences obtained in simulation results between a model in which independent and identically distributed (i.i.d.) assumptions are adopted and one in which autocorrelation effects are considered.

Serial production line performance under random variation: Dealing with the ‘Law of Variability’

“…These conclusions are also shared by Takahashi and Nakamura [51], Altiok and Melamed [52] and Pereira et al [53], who found that both positive and negative Corr(A) and Corr(S) have a negative impact on the performance of serial production lines.…”

Modelling and analysis of the impact of correlated inter-event data on production control using Markovian arrival processes

“…They show that positively correlated arrivals spent more expected time in the system. Pereira et al (2012) evaluate the impact of correlation on assembly processes, serial lines and disassembly processes. They demonstrate that considering autocorrelation in modeling the process enhances the accuracy of estimated performance measures of the real system.…”