Abstract
Traditional simulation modelling focuses upon the analysis of steady-state data. This focus may not be appropriate, however, for the study of transient responses – data reflecting some form of disruption or change in the system norms. Transient responses are often encountered when dealing with new product introductions, changes in production systems, or supply chain disruptions. In these situations, it is the transient response, how the system responds to these changes as well as the tactics and strategies used to deal with these changes, that tend to be of the greatest interest. Unfortunately, current approaches that focus on analysing such responses are limited. This paper introduces a new approach for analysing transient responses – one that merges outlier detection, a time series analysis tool, with simulation modelling. This combined approach allows the researcher to identify those factors that have the greatest impact upon operations during these transient conditions. Using a simulated supply chain disruption to illustrate the potential of the approach, it is shown that the new approach expands the applicability of simulation and enables certain types of problems to be investigated with confidence not previously provided.
Acknowledgements
The authors would like to express their thanks and appreciation to William Lattyak of Scientific Computing Associates Corporation (525 N. Lincoln Avenue, Villa Park, IL) for all of his valuable assistance and suggestions in this research project.