KEYWORDS: Ship traffic, Distribution, Data analysis, Risk analysis

Abstract

Simulation models have been used extensively in planning and analysis of port operations. Many different simulators exist, varying in complexity and objectives; some studying bulk terminals (UNCTAD 1969; Park and Noh 1987) and others studying container (Ramani 1996; Kozan 1996) or military (Nevins et al. 1998) terminals. However, there is a lack of detail in the published papers due to the fact that most of the models are proprietary (Holguin-Veras and Walton 1995). Moreover, the traffic data on which these models are based are also proprietary. As a result, each time a new simulation model is created, a new analysis of port operations and traffic data has to take place for model calibration and verification. In cases where simulation is performed in the planning stage, there is no data available with regards to ship specifications and distributions of loaded and discharged cargo as well as service time distributions. In this paper, the authors present the methodology that was followed in the analysis of port traffic and operations data for a particular simulation model that was created in the context of risk analysis. Although particular parameters are not presented here (and would pertain only to a specific port), the approach that is described is helpful in guiding the early stages of ship traffic and ship-berth operations modeling.

 
In the following sections the authors present the context for which the simulation model was needed, in order to clarify why this particular approach was followed. Next, the format of the data is described and the methods used to obtain the necessary information are summarized. In the main part of the paper the authors present the methods and discuss their findings with regards the types of distributions used to model ship inter-arrival times, container cargo distributions, ship lengths and drafts, crane assignment rules and terminal service rate. Eventually, validation with existing data shows that the modeling choices were quite successful for the purposes of the particular simulation model.

This project is partially supported by the Pacific Earthquake Engineering Research Center through Grant No. SA1831JB. The authors also wish to thank the Port of Oakland for providing them with the data that was analyzed in this paper.