The Calibration of PRIOR, A Computer System for the Simulation of Port Operations Considering Priorities

As part of the research conducted at The University of Texas at Austin on the implementation of priority systems at container ports, a simulation system was developed. This simulation system, PRIOR, has the unique capability of performing micro-simulations of port operations considering different operational schemes depending on the container's priority. This report focuses on describing the calibration process followed by the research team to ensure an adequate representation of the test case. Two different approaches to calibration were followed in this research, combined models and empirical service time distributions. In the former.case, the models estimate service time as a function of the tasks' attributes (e.g., distance travelled by the yard crane), the mathematical expression of the service process, and the set of parameters obtained empirically. This approach allows specific consideration of both systematic (Le., explained by the independent variables) and random components of the service time. The parameters of the models representing the systematic component of the service time were estimated using multiple regression. On the other hand, empirical distributions were used when the process' characteristics were not suitable for analytical modelling, such as gate processes.