An Experimental Study on the Impact of Network Segmentation to the Resilience of Physical Processes

The fact that modern Networked Industrial Control Systems (NICS) depend on Information and Communication Technologies (ICT) is well known. Although many studies have focused on the security of NICS, today we still lack a proper understanding of the impact that network design choices have on the resilience of NICS, e.g., a network architecture using VLAN segmentation. In this paper we investigate the impact of process control network segmentation on the resilience of physical processes. We consider an adversary capable of reprogramming the logic of control hardware in order to disrupt the normal operation of the physical process. Our analysis that is based on the Tennessee-Eastman chemical process proves that network design decisions significantly increase the resilience of the process using as resilience metric the time that the process is able to run after the attack is started, before shutting down. Therefore a resilience-aware network design can provide a tolerance period of several hours that would give operators more time to intervene, e.g., switch OFF devices or disconnect equipment in order to reduce damages.