In a world where more and more smart devices communicate with one another, industrial organizations are recognizing that a seamless flow of information created by connecting control systems to the enterprise is critical for making significant operational improvements.
A control system evaluates the condition of input variables (sensors), combined with requests from a human-machine interface, executes a logic or control sequence and drives some outputs (actuators).
The key difference is that any interference, including an attack to the network, in this process can affect the availability and quality of the manufactured product and can also pose a risk to the people, the facilities or the environment.
Corporate information systems have requirements regarding reliability, performance and risk management that are very different to those of an industrial control system. While corporate systems are focused on high performance, large amounts of data and scheduled tasks, the control system is in real-time, the process is continuous and the data volume is comparatively much lower.
While connecting devices helps create a seamless flow of information, protecting these industrial assets from security risks becomes increasingly more important. Corporate information systems rely on information protection. The unavailability of the system suggests potentially being able to stop the business operations, and a temporary service interruption can be acceptable, with recovery times around minutes or even hours.
However, in industrial control systems, the protection of people and the environment is key. The unavailability of the system can cause potential defects or damage to the manufactured product or the facilities, as well as a hazard for people or the environment, and, lastly, fault tolerance and/or very short recovery times are essential.
Therefore, a key point will be categorizing the facility to state what is the possible impact of a security problem that challenges the availability, integrity and confidentiality of the system.
The impact will not be the same if automotive components or hazardous chemical products are manufactured. Similarly, the sort of industry is relevant: food, pharmaceuticals or drinking water distribution. However, industrial security is important for all industries and applications. It requires a defense-in-depth security approach that addresses both internal and external security threats. Defense-in-depth security is a layered approach focusing on physical, network, computer, application and device security.
In order to carry out the categorization, several issues must be considered, such as the hazardousness of materials, production process, intellectual property, whether the manufactured products are intended for human consumption, potential impact on the environment, whether it's a critical infrastructure, etc.
Therefore, we should think about the defense-in-depth approach and acknowledge that we need well-defined security policies to build a more secure, industrial control network, with clear goals and procedures.
The personnel involved in production, maintenance, engineering, and IT must be aware of these policies and help ensure its compliance, so the corresponding information will be a must, as well as training about procedures to follow.
Sometimes the security problems experienced by industrial control networks are generated by actions taken from the operations and maintenance staff – in-house or external. For example, in an unsecure network with vulnerabilities in the physical layer, an employee could create a loop in the network and a potential traffic saturation, an inadvertent topology change or an accidental configuration change.
Therefore, the human factor is very relevant because it can initiate vulnerability, even unintentionally. Remember to take a comprehensive approach to security, because control systems, networks and software can all help defend against security threats and risks.
To learn more about network security, please visit our web site.
Published 24 February 2015
