Conceptual framework for container-handling process analytics

Abstract

Port logistics systems exhibit increasingly dynamic processes, as, in the course of container-handling activities, many changes of data occur and uncertain conditions arise. Frequent changes and uncertain conditions incur high cost and inefficiency in a port terminal process. In order to find the causes of these problems, in order, that is, to minimize costs and increase efficiency, the event logs of a port logistics process are explored. The first step herein is discovering a process model in relation to which the real-world processes can be understood. Analysis of the process model, subsequently, will reveal the causes of high cost and inefficiency. In the present study, we built a conceptual framework for application of process mining techniques to the domain of port logistics. Existing process mining techniques cannot be directly applied to our problem, for several reasons. First, there exist various resource types, such as truck, equipment and vessel. Second, careful pre-processing is required, since existing event logs do not provide for the concept of case id. Third, a domain-specific performance index needs to be considered. The new conceptual framework introduced in these pages was built purposely to provide guidelines for operators and other stakeholders who need to analyze the causes of inefficiency in port logistics processes. Later in this paper, we present a case study demonstrating the utility of our framework.