// OT AND ICS SECURITY · FAQ · HANS STUDY
OT security questions, answered
These are the questions operators, integrators, and IT teams ask most often about industrial control system security. Each answer is short, with a link to the longer treatment where one exists.
OT basics
What is OT security?
OT security protects the systems that run physical processes, such as PLCs, HMIs, SCADA servers, and the networks that connect them. The goal is to keep the process safe and running while limiting who and what can change it. It covers network design, remote access, patching, monitoring, and recovery planning, with the plant's safety and uptime constraints built in from the start.
What is the difference between OT and IT security?
IT security protects data first, while OT security protects safe, continuous operation first. IT teams can usually reboot a server overnight; an OT team may not get a maintenance window for months. Patching, scanning, and incident response all change shape as a result, which is why IT controls need testing before they touch a control network. The OT and IT convergence advisory covers the organizational side.
What is the difference between ICS, SCADA, and DCS?
ICS is the umbrella term, and SCADA and DCS are two common kinds of industrial control system. SCADA typically supervises geographically spread assets such as pump stations or substations, while a DCS controls a single-site process such as a treatment plant or a refinery unit. PLCs and RTUs sit under either one and do the direct control work.
Why can't OT systems be patched like IT systems?
Many OT devices run continuously, need vendor approval before any change, and may run software that the vendor no longer supports. A patch can also change timing or behaviour in a way that affects the process. Where patching is not possible, the usual answer is compensating controls such as segmentation, restricted remote access, and monitoring.
How do attackers usually reach OT systems?
The most common paths are through the corporate IT network, through remote access that was set up for convenience, and through internet-exposed devices. Vendor connections and shared credentials add to that. Joint guidance from CISA, the EPA, and the FBI lists reducing public internet exposure and changing default passwords among its first recommended actions. The controls that close those paths are covered in security controls for OT networks that hold up in production.
Standards (IEC 62443, NIST 800-82)
What is IEC 62443?
IEC 62443 is a series of international standards for the security of industrial automation and control systems. It covers the whole chain, from the asset owner's security program to the integrator's work to the design of individual components. It is a framework you adopt in parts, not a single checklist.
Which part of IEC 62443 applies to me?
It depends on your role. Asset owners mostly use 62443-2-1 for the security program and 62443-3-2 for risk assessment and zones and conduits. Integrators and service providers look to 62443-2-4 and 62443-3-3, and product suppliers to 62443-4-2 for component requirements.
What are the IEC 62443 security levels?
Security levels SL 1 to SL 4 describe the strength of attacker a zone is designed to resist. SL 1 covers casual or accidental misuse, SL 2 intentional attack with simple means, SL 3 intentional attack with sophisticated means and moderate resources, and SL 4 the same with extended resources. IEC 62443-3-3 defines the system requirements for each level, and the target comes from a risk assessment rather than a default.
What are zones and conduits?
A zone is a group of assets that share the same security requirements, and a conduit is the controlled communication path between zones. IEC 62443-3-2 uses them to turn a plant into pieces you can assess and protect separately. A small plant often ends up with a handful of zones, not dozens. There is a worked small-plant example.
Do I have to comply with IEC 62443?
Not by default, since it is a voluntary standard. It becomes a requirement when a regulation, a customer contract, or a procurement document points to it. Many owners use it as a design reference without seeking formal certification.
What is NIST SP 800-82?
NIST SP 800-82 is the US guide to securing operational technology, including ICS, SCADA, and DCS. It describes typical OT architectures, threats, and how to adapt security controls to OT constraints. It is guidance, not a certification scheme, and it ties its recommendations to the NIST SP 800-53 control catalogue.
Network segmentation and the Purdue model
What is the Purdue model?
The Purdue model is a reference architecture that places industrial systems in levels, from field devices and sensors at Level 0 up to enterprise systems at Level 4 and 5. Control sits at Levels 1 and 2, site operations and MES at Level 3, and business systems above. It gives teams a shared vocabulary for deciding what may talk to what. Real plants rarely match it exactly, as the Purdue model in practice versus on paper explains.
What is the industrial DMZ, or Level 3.5?
The industrial DMZ is a buffer zone between the enterprise network and the plant network, usually drawn at Level 3.5. It was not part of the original Purdue model and was added later as plants connected to IT systems. Data and remote sessions pass through servers in that zone instead of going straight into the control network.
How do I segment an OT network?
Start by mapping what talks to what, then group devices into zones by function and consequence and put a firewall at each boundary. Allow only the specific protocols and addresses each conduit needs, and deny the rest. Do it in stages, with the plant's operators involved, because a rule that blocks a legitimate flow can stop a process. The course Hardening Control Networks teaches the boundary design.
Should an OT network be air gapped?
A true air gap is rare today, and a network called air gapped often has a vendor modem, a USB workflow, or a laptop that crosses the line. A well-defended connection with a DMZ, one-way data flow where it fits, and tightly controlled remote access is usually easier to maintain and easier to verify. Whatever the design, test the claim by looking for the paths that exist.
What ports do common industrial protocols use?
Common defaults are Modbus/TCP on 502, DNP3 on 20000, EtherNet/IP on 44818/TCP and 2222/UDP, Siemens S7 on 102, OPC UA on 4840, and BACnet/IP on 47808/UDP. Vendors and sites can change them, so treat the list as a starting point for firewall rules and check it against the actual traffic. The free course covers the protocols in OT Networks Level 1.
Are Modbus and DNP3 secure?
Not by themselves. Standard Modbus/TCP has no authentication or encryption, so anyone who can reach port 502 can usually read or write values. DNP3 has an optional Secure Authentication extension that many deployments do not enable, and OPC UA includes security features that only help when they are configured. Segmentation and protocol-aware firewall rules do much of the protecting.
Remote access
How do I secure remote access to an OT network?
Route every remote session through one controlled entry point, such as a jump host in the DMZ, with multi-factor authentication and named accounts. Give each person access only to the systems their task needs, for the time they need it, and log the sessions. Remove always-on connections, shared passwords, and direct internet exposure of control devices.
How should vendors get remote access to OT?
Vendors should connect through the same controlled entry point as staff, with their own accounts, approval for each session, and logging. A standing VPN that nobody remembers is a common finding, so time-limit access and review it on a schedule. Contracts can also state who owns the access path and how it is closed. See vendor remote access to OT without the forgotten VPN.
Is a VPN enough for OT remote access?
No, a VPN only protects the connection, not what the user can reach once inside. A VPN that lands on the control network gives the remote user, or anyone with their credentials, broad reach. Terminate the VPN in the DMZ, require MFA, and use a jump host or similar broker to limit what the session can touch.
Asset inventory and monitoring
How do I build an OT asset inventory?
Combine what you already have, such as drawings, spare-parts lists, and vendor records, with passive observation of network traffic and a physical walk-down. Record the make, model, firmware, location, network address, and owner of each device. Keep it updated, because an inventory that is a year old is a historical document. The method is laid out in building an OT asset inventory without active scanning.
Why is an asset inventory the first step in OT security?
You cannot segment, patch, monitor, or recover what you do not know exists. CISA and its partner agencies published asset inventory guidance for OT owners and operators in August 2025 for that reason. An inventory also exposes end-of-life devices and unexpected paths into the plant.
Can I use Nmap or active scanning on an OT network?
Only with great care, and often not at all on a live control network. Some PLCs and older devices have crashed or stalled when scanned. Passive monitoring through a SPAN port or network tap avoids that risk, and any active work should be tested on a spare device or done in a maintenance window with the vendor's guidance.
What is passive OT network monitoring?
It is listening to a copy of network traffic, from a SPAN port or tap, to identify devices, protocols, and changes without sending any packets to the control network. It builds an inventory, a baseline of normal communication, and alerts for new devices or unusual commands. Because it only listens, it carries far less operational risk than scanning.
Utilities and critical infrastructure
What is NERC CIP and who must comply?
NERC CIP is a set of cybersecurity and physical-security reliability standards for the North American bulk electric system. They apply to registered entities such as generator owners and operators and transmission owners and operators, and each standard has its own applicability section. Whether a particular facility is in scope depends on its impact rating, so check NERC's materials and your regional entity. The critical infrastructure page describes the advisory work for utilities.
What should a water utility do first for cyber security?
Reduce exposure to the public internet, change default passwords, turn on multi-factor authentication where it is supported, and inventory your OT and IT assets. Those items come from the CISA, EPA, and FBI fact sheet on top cyber actions for water systems, which also covers backups, incident response plans, and regular assessments. Small systems can ask the EPA about its free cybersecurity help.
Does Canada have an OT security law for critical infrastructure?
Federal legislation exists but is not fully in force. Bill C-8 received Royal Assent on 15 June 2026 and includes the Critical Cyber Systems Protection Act, which comes into force by order in council. Sector and provincial rules can apply as well, so check what binds your operation today. See what the CCSPA means if you service critical infrastructure.
Where does a small team start with OT security?
Start with the inventory, then remote access, then segmentation of the highest-consequence processes, in that order. Each step reduces risk without needing a monitoring platform. A free introduction to the networking and protocol background is in the OT Networks Level 1 course, and the industrial OT networks advisory covers independent review before money is committed.
Have an OT network question that is not here? Book a scoping call.