A partition berth allocation scheduler based on resource utilization and load balancing

Abstract

The systematic solutions to a berth allocation problem (BAP) at container terminals have been scarce owing to their high complexity and strong randomness. This paper extends the past research on modeling of container terminal logistics systems (CTLS) with computational thinking. We introduce and integrate the task scheduling architecture and mechanism in real-time operating system (RTOS) into BAP, and propose a partition berth allocation scheduler based on resource utilization and load balancing. Subsequently, we establish the fundamental principles of scheduler with the computing perspective and decision framework of ARINC 653, which is a design philosophy of RTOS for integrated modular avionics. Finally, the approach is demonstrated by investigating the stress testing of a typical container terminal logistics service case in contrast with the average random berth assignment algorithm based on the comprehensive computational experiments.