Cyber intrusion detection in operations of bulk handling ports

Bulk handling ports are integral to the well-being of resource-driven export economies. They rely on Industrial Control Systems (ICS) to coordinate the receiving, stockpiling, processing and loading of bulk carriers to sustain the 24/7 movement of millions of tons of material. Defense of an ICS against cyber-attack is essential to the resilient operation of these ports, and identification and remediation of intrusions are significant components of effective defense practice. This paper investigates the role of Machine Learning (ML) algorithms in designing an efficient Intrusion Detection System (IDS) that can recognise malicious activities that confront the ICS of a bulk handling port and includes an assessment of existing IDS-based ML methods. We also design a bulk port cyber simulation to evaluate some well-known ML techniques and their capability for establishing an effective bulk handling port ICS IDS. The empirical results reveal that the use of ML techniques can achieve high performance in terms of accuracy and false alarm rates in a simulated environment. However, a comparison of a simulated bulk handling port environment with a production environment led us to conclude that our simulation needs further analysis and modelling in order to be applied in real Bulk handling ports.