Security and Data Protection for Future Production Systems


Security and Data Protection

This research field deals with sustainable research questions for secure industrial production plants. In addition to the legal aspects of data protection, these include future, flexible security concepts for industry 4.0, secure cloud use, self-learning anomaly detection in industrial production, and verifiable security in the presence of active adversaries.

Flexible Security Concepts

In Industry 4.0, intelligent autonomous components will increasingly be used in the future, which are to interact and communicate spontaneously with other, already integrated components. This requires greater flexibility of the networks and the predefined security zones.

The concept of software defined networks (SDN) offers the possibility to implement security guidelines for devices, applications and services in a detailed and flexible manner.

Smart Fabric can thus benefit from SDN as a building block for flexible and innovative security concepts. KASTEL develops concepts that enable the use of software-defined network technologies to implement modern security concepts in industry 4.0.

For example, for the dynamic establishment of security zones or the flexible composition of security-relevant network functions and their placement within the physical infrastructure.

Participating research groups are the Fraunhofer IOSB and the TeleMatics.

Cloud Computing

In the context of Industry 4.0, businesses are expected to use cloud computing technology for secure data storage and data exchange between companies. Cloud computing becomes part of a critical infrastructure for industry. The advantages of flexibility, robustness and cost savings are offset by the loss of transparency.

KASTEL is developing a framework to increase the transparency of cloud-based industry 4.0 solutions. This is intended to enable a company as a cloud user to check, for example, whether a solution actually complies with the requirements. For example, at which geographical locations the data is stored and whether the required redundant copies have also been created or deleted in accordance with the regulations.

The research group in this area are the TeleMatics.

Legal Aspects

Like all innovations, Industry 4.0 also encounters a legal environment that has to be taken into account in its development. The identified statutory and European legal requirements must be examined with regard to the specified application scenarios, whereby questions of data protection law in particular are of central importance. At European level, the new basic European data protection regulation (DS-GVO) must be taken into account, the standards of which will apply from May 2018. Their rules will apply directly in each Member State and will replace national data protection legislation in huge parts.

KASTEL is investigating how the computerisation of manufacturing technology can be promoted within the framework of Industry 4.0 in such a way, that data and secrecy-protecting precautions can nevertheless be taken.

The participating research group is the Center for Applied Law (ZAR).

Real-time Requirements and Verifiable Security

Security in plants that implement the Industry 4.0 concept must be considered comprehensively - from the planning level to the technical levels. The systems are operated in real time, which poses an additional security risk. An adversary who gains access to the technical infrastructure of a plant can cause great physical and financial damage.

KASTEL develops a formal method for conclusively demonstrating security.

Specifically, it is to be shown that an adversary with the means at his disposal is incapable to damage the plant or to operate it outside of the envisaged parameters. For this purpose, absolute properties (e. g."the drill head never moves deeper than expected into the drilling material") and relational properties (e. g."the speed of the motor can be at most doubled by reconfiguration") are to be investigated.

Research is carried out together with the Institute for Theoretical Computer Science (ITI).

Security and Data Protection for Future Production Systems

Modern production facilities are highly networked. Embedded systems communicate with each other independently, planning systems from the cloud calculate order steps and machine occupancy, plant operators monitor and control from a distance, maintenance personnel access resources worldwide and perform configuration changes. In the networked world, the protection of production facilities no longer ends at the factory building or the company grounds. The network connections allow adversaries to intrude and manipulate the systems, malware infections can completely paralyze large areas of the system, causing immense physical damage to the system and danger to the population. Not only since news about Stuxnet, Duqu, Flame and Havex has it been clear that production facilities are easy targets for cyber attacks.

Industry 4.0 is increasingly dissolving the previous separation of traditional IT networks and production networks in order to be able to operate communication and data exchange across all network hierarchies. Network components in production are clearly different from the components used in traditional IT. In its development, which is designed for a service life of several decades, networking and the associated data security have so far played little role. Historically, production lines are separated from each other and from other IT systems. This separation was enforced physically, by separate communication networks and also logically, by different protocols. In the course of Industry 4.0, these systems will now be connected to the network systems of traditional IT. The industry hopes that this will result in more flexible and efficient production processes. However, the production systems are also exposed to many hazard scenarios of traditional IT systems, which makes IT security an important aspect of industrial systems. In order not to jeopardise the success of industry 4.0, the use of new technologies must not become a security risk.