OT security and connected control systems – how to reduce the risks

OT security and connected control systems – how to reduce the risks

The attacks on refrigeration systems show how vulnerable connected control technology can be. Seven areas to check: inventory, exposure, segmentation, remote access, vulnerabilities, monitoring and recovery.

Summary: The cyberattacks on connected refrigeration and freezer systems in Swedish and European shops in the autumn of 2026 show how vulnerable technology that controls physical processes can be. Similar systems exist in buildings, industry, warehouses, water and wastewater plants and many other operations. Many of them have a long service life, limited ability to take security updates and remote connections that are hard to keep track of. This article covers how organisations can improve their OT security through inventory, network segmentation, protected remote access, vulnerability management and working incident readiness, without putting operations at risk.

The attacks on refrigeration systems point to a bigger problem

Since 11 October 2026, refrigeration and freezer systems and similar connected devices in several EU countries have been hit by cyber incidents. Swedish shops have been affected too. On 2 November the Swedish National Cyber Security Centre (NCSC) announced that it is following the events and urged organisations to review the security of their OT environments. On 5 November ENISA and the EU’s joint cybersecurity networks issued a joint statement after a new wave of incidents, and on 11 November the Swedish government published a reinforced call to review cybersecurity, particularly where older technology is connected to the internet.

It is easy to see the attacks as a problem for grocery retail. But the underlying question is much bigger: how many organisations really have full control of the technology that runs their physical operations? It may be ventilation, cooling, heating, access control, production equipment, pumps, water treatment or other building and industrial automation.

The systems may have been installed ten or twenty years ago and worked without major problems since. Meanwhile the conditions have changed. More systems have been connected to networks and the internet for remote control, monitoring and more efficient service. That creates new opportunities, but also security risks the organisation needs to understand and manage.

What is OT and how does it differ from ordinary IT?

OT stands for operational technology: technology that monitors and controls physical processes, machines and plants. Examples of OT are:

  • control systems for ventilation, heating and cooling in buildings,
  • PLCs and industrial control systems in production,
  • SCADA systems for monitoring and control,
  • control of pumps and of water and wastewater plants,
  • building automation and other connected technical installations.

In traditional IT the focus is often on information, users, applications and system availability. In OT the physical process itself can be affected. A security problem can stop a production line, change the temperature in a facility or leave technical functions impossible to control as intended. OT security therefore has to take cybersecurity, operational reliability and, in some operations, personal safety into account.

That also affects how the security work is carried out. A server can often be updated or restarted in a planned maintenance window. A control system that keeps a critical process running can be much harder to switch off.

Why are connected control systems particularly exposed?

The NCSC describes OT as a particularly exposed part of organisations’ technology environments. The long service life means many systems are outdated, updates can cause production outages, and some older protocols lack basic security controls. At the same time it is common for external suppliers to need to connect for service and troubleshooting.

That raises questions the organisation needs to be able to answer. Who can connect to the equipment, and how is the user authenticated? Which other parts of the network can be reached from there? Is there someone responsible who knows which components are in use and which security updates are missing? And who decides when a vulnerability needs to be fixed but the update carries a risk of downtime?

The problem is not always that the technology lacks security features. Sometimes the features exist but have never been enabled, documented or followed up.

Seven areas to check in your OT environment

1. Inventory the systems and find out what is actually connected

The first step is a reliable picture of the environment. Which control systems exist, where are they installed, what do they do and who is responsible for them? The inventory also needs to cover network equipment, operator panels, gateways, monitoring servers, integrations and remote connections. For each relevant system you should be able to state the manufacturer, model, software version, responsible supplier and whether the equipment still has security support.

An equally important question is how the equipment is connected: on an internal network, in a separate OT network, behind a supplier’s router or directly reachable from the internet? Without an up-to-date inventory it is hard to carry out a relevant risk analysis or prioritise measures.

The inventory must also be done without disturbing operations. Active network scanning that works well in an office environment can affect sensitive OT equipment. The mapping should therefore be planned together with system owners and the suppliers concerned, using document review and passive methods where appropriate.

2. Limit exposure to the internet

A control system does not need to be reachable from the internet to be administered or monitored. The NCSC’s first piece of advice after the events is to inventory which devices are connected to the internet. Then assess whether each connection is really needed, and review public IP addresses, open ports, remote administration, supplier connections and cloud links.

If an internet connection is needed, it should be protected with appropriate technical controls, and administration interfaces should not be exposed directly to the internet when a more secure access solution is available. But an established connection should not be closed before you understand what it is used for either. That could affect monitoring, alarms or other operationally critical functions.

3. Segment the network

A common weakness is that control systems, user computers, servers and other equipment have unnecessarily broad access to each other. If an office system is compromised, the attacker should not be able to reach the control systems without further ado. Network segmentation divides the environment into security zones with controlled communication between them, for example:

  • ordinary office IT,
  • production and control systems,
  • building automation and IoT,
  • system administration,
  • external supplier connections.

The aim is not to create as many VLANs as possible. What matters is defining which systems need to communicate, which protocols and ports are needed and which traffic should not be allowed. A separate VLAN does not in itself mean the traffic between zones is sufficiently restricted; segmentation has to be combined with firewall rules and access controls, and traffic flows need to be documented and followed up. How segmentation is built in practice is covered in our insight on network segmentation for businesses.

4. Review suppliers’ remote access

Many OT systems are maintained wholly or partly by external suppliers: machine manufacturers, automation firms, building suppliers or specialists in particular parts of the technology. Remote access may be necessary for effective service, but the organisation needs to know how it works. Does the supplier use its own account, or do several technicians share the same administrator account? Is multi-factor authentication used where it is technically possible? Is the supplier limited to the systems it is responsible for, or can the connection be used to reach other parts of the network? Can the access be closed when it is not needed?

A practical improvement is to bring remote access into one controlled solution with individual identities, clear permissions, logging and time-limited access. Also document who in your own organisation is responsible for the supplier’s access and follows it up.

5. Manage vulnerabilities without creating new operational problems

Vulnerability management in OT often needs a different approach from ordinary IT. There may be known weaknesses that cannot be fixed by installing an update straight away: the supplier may no longer provide updates, or the change first has to be verified so that the control system keeps working. That does not mean the risk should be left untreated. Planned updates may need to be combined with segmentation, restricted remote access, reinforced monitoring and other compensating measures.

What matters is that the risk is known, assessed and assigned to someone responsible. A critical vulnerability in an internet-exposed control system normally needs a different priority from the same vulnerability in an isolated environment with several layers of protection. Updates and technical tests are therefore planned together with the business and the suppliers, taking into account operations, safety and the ability to restore the system if the change does not work.

6. Ensure monitoring and the ability to detect anomalies

A control system that works is not necessarily secure. A system can keep delivering its function while someone unauthorised has gained access to its administration or changed settings. Assess which events should be detectable: new network connections, unexpected logins, changed configurations, new devices or communication that deviates from normal. OT monitoring may also need to combine technical security events with information from the process itself, such as an unexpected change in a control value or repeated restarts.

Monitoring has to match what the systems can cope with and which data can actually be collected. And someone needs to receive and assess the alerts. A tool that collects information is not the same as a working incident management process.

7. Prepare incident management and recovery

What happens if the organisation loses access to a critical control system? Can the process continue at reduced function, and is local control of the equipment possible? Are configurations, control programs and other recovery material backed up? Who contacts the supplier, and who may decide that a system is to be isolated or stopped?

For OT, incident planning has to take the consequences for the physical operation into account. Disconnecting a system may be right from a cybersecurity perspective but at the same time create risks for production, property or people. Decisions on isolation and recovery therefore need to be made together with people who understand the technical process. Also test the recovery capability: a backup is not enough if the organisation lacks the documentation, equipment or expertise needed to restore the control itself. How business requirements, dependencies and recovery capability fit together is covered in Business continuity management in practice.

An example: the building system that became part of the corporate network

Picture an organisation with several buildings. Ventilation, heating and cooling are controlled through a central system, and an external supplier can connect remotely for troubleshooting and service. The system was installed many years ago. Since then more functions have been connected to the network, and some parts have been integrated with the ordinary IT environment. Everything works as it should.

A security review then shows that the remote access has not been followed up for several years, that the network rules allow more communication than necessary and that the documentation no longer describes how the systems are actually connected. That does not have to mean anyone made an obvious mistake. The environment has evolved step by step through projects, supplier changes and technical improvements, but the overall risk picture has changed.

A relevant action plan can then start by mapping the systems and the access, restricting communication between network zones, reviewing the supplier’s permissions and making sure there is a working incident and recovery process. The work can be done in stages, dealing with the riskiest conditions first.

The Swedish Cybersecurity Act, the CRA and ISO 27001: how do the requirements fit together?

For organisations affected by the Swedish Cybersecurity Act (NIS2), risk management, supply chain security, incident management and continuity are central parts of the security work, and that can include control systems and other operational technology where they form part of the organisation’s network and information systems. Not every organisation that uses OT is automatically affected by the Act, though. That depends on the organisation’s sector, size and the Act’s other scope rules.

The Cyber Resilience Act (CRA) has a different focus. It sets cybersecurity requirements for manufacturers and other economic operators that make products with digital elements available on the EU market. Since 11 September 2026 manufacturers must report severe incidents and actively exploited vulnerabilities, and most of the product requirements apply from 11 December 2027. For organisations that use connected control systems, the CRA is therefore particularly relevant when procuring new equipment and assessing suppliers’ security work. It does not mean the organisation itself takes on manufacturer responsibility.

ISO 27001 can give structure to risk assessment, asset management, supplier governance, security measures and continual follow-up. But neither regulations nor standards replace the need to understand what the technical environment actually looks like.

Where should you start?

A large OT security programme is not always needed to make progress. A good first step is to map the most critical connected technology and carry out a focused review of the systems’ exposure, network connections, remote access and allocation of responsibility. Then you can separate the risks that need handling immediately from the improvements that can be planned with the business and the suppliers.

The result should be concrete: a documented picture of the current state, identified risks, proposed technical and organisational measures and a clear prioritisation. A review is not finished just because the weaknesses have been identified. Someone also needs to be responsible for making sure the changes are carried out and followed up.

How Kristensson i Skåne can help

Kristensson i Skåne helps organisations with information security, cybersecurity and technical security, as part of our work on cybersecurity and technical security. For organisations with connected control systems we can support with mapping networks and system dependencies, reviewing segmentation, firewall rules and remote access, and risk assessment with a prioritised action plan. We also help with supplier governance, incident readiness, business continuity and linking the work to the organisation’s existing management system or regulatory requirements.

When a review touches sensitive production or control systems, the work is planned together with the organisation’s system owners and the relevant technology suppliers. Any active tests are scoped and carried out with the right OT expertise and safety conditions. The starting point is that security should improve without creating unnecessary risks for the operations the technology is there to support.

Frequently asked questions

What is the difference between IT security and OT security?

IT security is mainly about information, users and systems. OT security concerns technology that controls physical processes, so a security problem can also affect production, temperature, flows or personal safety. OT security therefore has to account for operations and the process, and measures such as restarts or active scanning cannot always be carried out the same way as in IT.

What should we do first after the attacks on refrigeration systems?

Inventory which devices are connected to the internet, as the NCSC recommends, and check whether your refrigeration, freezer or other control systems are covered by the current security advisories. Then assess whether each internet connection is needed and how suppliers’ remote access is protected.

Can we scan the OT network with the same tools as the office network?

Not without planning. Active scanning that works in an office environment can affect sensitive OT equipment. The mapping should be planned with system owners and suppliers and rely on document review and passive methods where appropriate.

Are we affected by the Swedish Cybersecurity Act because we have control systems?

Not automatically. Whether the Act applies depends on the organisation’s sector, size and the Act’s scope rules. If it does apply, the security work covers the network and information systems the organisation uses, which can include control systems.

What does the CRA mean for us as buyers of connected equipment?

The CRA sets requirements for manufacturers and other economic operators, not for you as a user. The reporting obligation for manufacturers has applied since 11 September 2026 and most of the product requirements apply from 11 December 2027. For you, the CRA is a support when procuring and when assessing suppliers’ security work.

Want to know how secure your connected systems and networks are? Read more about our offer vulnerability assessment and penetration testing or contact us for a first conversation about your environment, your suppliers and which measures matter most.

Sources and further reading: Swedish National Cyber Security Centre, Cyberangrepp mot uppkopplade frys- och kylsystem, 2 November 2026; NCSC, EU varnar för cyberincidenter i kyl- och fryssystem, 5 November 2026; ENISA, Joint statement on cyber incidents in refrigeration systems across the EU; Government Offices of Sweden, Förstärkt uppmaning om att se över cybersäkerheten, 11 November 2026; NCSC, Operativ teknik – rekommendationer för beslutsfattare och organisationer; NIST SP 800-82 Rev. 3, Guide to Operational Technology (OT) Security; Regulation (EU) 2024/2847 (CRA). Verified 12 November 2026.

This article gives an overview of security risks and measures for operational technology. Which protective measures and regulatory requirements are relevant has to be assessed on the basis of the organisation, the equipment and the technical conditions. Active security tests and changes in OT environments should always be preceded by an assessment of the operational and safety risks.