ISA/IEC 62443Risk AssessmentSpecialist study guide Open the practice exam

62443 Risk Assessment Specialist Study Guide

Condensed, original study notes for the ISA/IEC 62443 Cybersecurity Risk Assessment Specialist exam (course IC33), with a matching practice exam in this repository. Not affiliated with or endorsed by ISA or IEC.

On this page
  1. Exam at a glance and a 7-day plan
  2. 1 · Planning and scoping an assessment
  3. 2 · Asset inventory, diagrams and data flows
  4. 3 · Vulnerability assessment and testing
  5. 4 · Threats and threat scenarios
  6. 5 · Risk criteria: scales, matrix and tolerable risk
  7. 6 · Initial risk assessment and partitioning (ZCR 1–4)
  8. 7 · Detailed risk assessment (ZCR 5)
  9. 8 · Process safety links: PHA, cyber PHA and the SIS
  10. 9 · Countermeasures, SL-T and residual risk
  11. 10 · The CRS, reports and approval (ZCR 6–7)
  12. Exam traps, tie-breakers and sources

Exam at a glance and a 7-day plan

ItemDetail
CertificateISA/IEC 62443 Cybersecurity Risk Assessment Specialist
CourseIC33, Assessing the Cybersecurity of New or Existing IACS Systems (self-paced: IC33M)
Format90 multiple-choice questions, closed book
Time2 hours
DeliveryMeazure Learning test center, or online with live remote proctoring
ResultPass or fail. ISA does not publish the passing score
PrerequisiteThe Fundamentals Specialist certificate
Counts towardThe ISA/IEC 62443 Cybersecurity Expert designation, with Fundamentals, Design and Maintenance

This exam is about carrying out an assessment. ISA's published objectives cover scoping and documenting the system, gathering information, finding vulnerabilities, interpreting process hazard analysis (PHA) results, facilitating a risk assessment, judging threat scenarios and existing countermeasures, finding gaps in policies and procedures, weighing the cost and effectiveness of new countermeasures, defining zones and conduits, and writing a cybersecurity requirements specification (CRS).

Most of that hangs on the 62443-3-2 workflow, so learn its zone and conduit requirements (ZCR 1 to 7) in order, with what each step needs and what it produces. Expect scenario questions that ask what to do next, which input is missing, or which document holds a given fact.

DayReadPractice
1Planning and scope; inventory and diagramsQuick 20 for a baseline, then domains 1–2
2Vulnerability assessment; threatsDomains 3–4
3Risk criteria; initial assessment and partitioningDomains 5–6
4Detailed risk assessment; process safety linksDomains 7–8
5Countermeasures and SL-T; the CRS and reportingDomains 9–10
6Nothing newFull 90-question mock under time, then review misses
7Traps and tie-breakers; your two weakest domainsSecond mock. Aim for 80% or better in every domain

Original study notes, not ISA course material. Not affiliated with or endorsed by ISA or IEC. Check clause numbers against your licensed copies of the standards.

1 · Planning and scoping an assessment

Every assessment starts with agreement on what is being assessed, why, by whom and under what rules.

Define the system under consideration (SUC). 62443-3-2 starts here (ZCR 1): draw the boundary of the system and list every access point into it, such as network connections, remote access paths, wireless links, removable media and maintenance ports. Anything outside the boundary is treated as an external interface, described by assumptions.

Planning itemWhat to settle
ObjectiveNew design or existing system; high-level or detailed; risk assessment or gap assessment against a standard
ScopeSites, systems and lifecycle stage covered; what is excluded and why
SponsorThe asset-owner manager who approves the scope and will accept the results
TeamA facilitator plus operations, process and control engineering, process safety, OT and IT security, maintenance, and vendors or integrators where needed
Rules of engagementWhat may be inspected, scanned or tested, when, by whom and with which tools; approvals; stop conditions; emergency contacts
Information requestArchitecture and network diagrams, asset inventory, policies and procedures, PHA and safety studies, firewall configurations, earlier assessments, incident history
Handling of resultsWho may see findings, and how they are stored and shared
DeliverablesReport, risk register, zone and conduit diagrams, and the CRS

Why a cross-functional team? No single person knows both the process consequences and the network. Operations and safety staff know what a compromised function could do; engineers and security staff know how it could be reached.

Typical flow of an assessment

  1. Pre-assessment: agree scope and rules, send the information request, review documents and questionnaires.
  2. On site: interviews, walkdowns of control rooms, cabinets and network closets, observation of practices, and passive data collection.
  3. Analysis: vulnerabilities, threats, consequences and likelihood combined into risk.
  4. Reporting: findings, risk ranking, recommended countermeasures, and requirements for the CRS.

New or existing system? For a new system, assess during design, when changes are cheapest. For an existing system, first establish the current state, then compare it with the risk criteria and the target. Reassess periodically and after significant changes.

Assessment results map a site's weaknesses, so treat them as confidential: share on a need-to-know basis and protect stored copies.

Exam angle: identify the SUC before anything else; the asset owner approves scope and accepts results; include operations and safety staff; agree rules of engagement before any active testing.

2 · Asset inventory, diagrams and data flows

You cannot assess what you have not found. The inventory and diagrams are the foundation for every later step.

What to record for each asset

GroupFields
IdentityName or tag, type (PLC, HMI, server, switch, firewall, RTU, SIS controller), manufacturer, model
VersionsHardware revision, firmware, operating system, applications and patch level; vendor support status
ConnectivityIP and MAC addresses, serial and wireless interfaces, protocols and services in use, remote access
ContextLocation, zone, owner, function, criticality, accounts, backup status

How to build it

  1. Start from documents: drawings, purchase records, project and configuration files.
  2. Confirm with walkdowns and interviews: what is actually in the cabinets and on the network.
  3. Use passive network monitoring to discover devices and flows without sending traffic.
  4. Use active queries or scans only where they are approved and known to be safe for the devices.
  5. Reconcile: every undocumented device or connection is a finding in its own right.

Diagrams the assessor needs

Critiquing an architecture diagram means looking for patterns like these:

PatternWhy it matters
Flat network with no internal boundariesOne compromise reaches everything
Business systems connected straight to control, with no DMZNo place to broker or inspect traffic
Dual-homed hostsBridge zones around the firewall
Remote access landing inside the control networkThe remote endpoint becomes an insider
Safety system sharing the control networkA control-side compromise can reach the SIS
Unmanaged modems, wireless or cellular gatewaysHidden external access points
Shared services such as domain controllers or antivirus servers spanning zonesTrust that crosses boundaries

Cyber criticality rates assets or functions by the consequence of their compromise, such as safety, control, monitoring or business-only. It sets how deeply each part is assessed and helps group assets into zones.

Exam angle: documents first, then verification; passive discovery before active; undocumented connections are findings; data flows drive conduit rules; keep the inventory current through management of change.

3 · Vulnerability assessment and testing

A vulnerability is any weakness a threat could exploit. In an IACS assessment, look well beyond software bugs:

MethodHow it worksRisk to a running process
Document and policy reviewCompare policies, procedures and designs with a standardNone
Interviews and walkdownsTalk to staff; inspect rooms, cabinets, ports and labelsNone
Configuration reviewExamine exported firewall rules, switch and host settings, account listsNegligible
Passive traffic captureCopy traffic from a SPAN port or network tap into an analyzer such as Wireshark to see devices, protocols and flowsNegligible if the tap or SPAN is set up carefully
Host checksReview patches, services, accounts, shares and logging on Windows hostsLow, with approval
Active scanningPort and vulnerability scanners such as Nmap, Nessus or GFI LanGuardMedium to high: fragile devices can fault
Penetration testingExploit weaknesses to prove impact, for example with Kali Linux toolsHigh: test systems or approved windows only

A gap assessment compares current practice with a standard, such as 62443-2-1 or 62443-3-3, or with company policy, and lists the gaps.

CSET, CISA's free Cyber Security Evaluation Tool, guides a structured, questionnaire-based assessment against selected standards, lets you draw the network, and reports the gaps. It does not scan anything, so its results are only as good as the answers it is given.

Vulnerability information sources: vendor advisories, CISA ICS advisories, the National Vulnerability Database (CVE identifiers with CVSS scores), and CISA's Known Exploited Vulnerabilities (KEV) catalog.

CVSS base scoreSeverity
0.1–3.9Low
4.0–6.9Medium
7.0–8.9High
9.0–10.0Critical

CVSS rates technical severity, not the risk to your plant. Exposure, consequence and existing countermeasures still decide priority, and a medium-scored flaw on a reachable safety-related host can outrank a critical one on an isolated test bench.

Safer active testing: use a lab, spare equipment or a staging system where you can; otherwise test during a planned shutdown with operations' approval. Throttle scans, exclude known fragile devices, watch the process during testing, agree stop conditions, and have vendor support and recovery plans ready.

Exam angle: passive before active; CVSS is not risk; CSET is a questionnaire-based standards assessment; penetration tests belong on representative test systems.

4 · Threats and threat scenarios

A threat scenario is the unit of a detailed risk assessment. It links five things:

ElementQuestionExample
Threat sourceWho?A contractor whose laptop carries ransomware without the contractor knowing
Vector or access pointHow do they get in?Plugging the laptop into the engineering network
VulnerabilityWhat weakness is used?Flat network with open file shares and no allowlisting
TargetWhich asset or function?Operator HMIs
ConsequenceWhat happens?Loss of view; the unit is shut down by hand; days of lost production

Threat sources differ in capability and motivation, which is also how the security levels describe attackers:

SourceTypical capabilityClosest SL
Careless or untrained insider, untargeted malwareNo intent; ordinary toolsSL 1
Opportunist, hacktivist, disgruntled insiderIntent, but simple means and generic skillsSL 2
Organized group or insider with IACS knowledgeSophisticated means, IACS-specific skills, moderate resourcesSL 3
Nation-stateSophisticated means, extended resources, high motivationSL 4

Vectors to enumerate come straight from the SUC's access points: the business network, remote access and vendor connections, internet-exposed services, wireless, removable media and visiting laptops, the supply chain (software updates and firmware), physical access to cabinets and ports, and insiders.

Threat information sources: CISA advisories, sector ISACs, vendor security bulletins, published incident analyses, and MITRE ATT&CK for ICS to check which adversary techniques your scenarios cover.

Structured ways to find scenarios

Judging a scenario: is it credible, meaning the source has the capability, motivation and access? Which existing countermeasures would stop or detect it? What is the worst credible consequence? Aim for realistic scenarios, not the worst imaginable one.

In 62443-3-2, identifying threats is the first step of the detailed assessment (ZCR 5.1): for each zone or conduit, describe the threat sources, their capability, the vectors they could use and the assets affected.

Exam angle: a scenario needs a source, a vector, a vulnerability, a target and a consequence. Build vectors from the SUC's access points, and test each scenario for credibility.

5 · Risk criteria: scales, matrix and tolerable risk

Agree how risk will be measured before assessing it. The asset owner's risk criteria set the scales, and the tolerable risk they define is recorded in the CRS (62443-3-2 ZCR 6.8).

Consequence scale. Each company defines its own; this one is only an illustration. Use the same categories and levels as process safety and enterprise risk management, so a cyber scenario can be compared with any other risk.

LevelPeopleEnvironmentProduction and costReputation
1 NegligibleNo injuryNo releaseUnder an hourNot noticed
2 MinorFirst aidContained on siteHoursLocal
3 ModerateLost-time injuryReportable releaseA shift or a dayRegional
4 MajorPermanent injuryRelease off siteDays to weeksNational
5 SevereFatalityMajor lasting damageMonthsInternational

Likelihood scale. For cyber events there is little reliable frequency data, so likelihood is usually judged qualitatively from the threat's capability and motivation, how exposed the target is, how easily the vulnerability can be exploited, and which countermeasures stand in the way. Some methods score attack difficulty instead, which works the other way round.

Risk matrix. Likelihood against consequence gives a risk rank, often low, medium, high and very high. The owner marks which cells are tolerable, which need treatment, and which are intolerable whatever the cost. Some companies borrow the safety idea of reducing risk as low as reasonably practicable (ALARP) for the band in between.

MeasureHow it is expressedNotes
QualitativeWords such as low, medium, highFast and common in OT; can be subjective
Semi-quantitativeScores on scales, multiplied or addedEasier to rank; still judgment-based
QuantitativeMoney and frequency, such as ALE = SLE × ARONeeds data that is rarely available for cyber events

Unmitigated and residual risk. 62443-3-2 first rates the unmitigated likelihood and risk, ignoring existing countermeasures (ZCR 5.4 and 5.5). It then counts the countermeasures already in place and re-rates likelihood and impact (ZCR 5.8 and 5.9) to find the residual risk (ZCR 5.10). Comparing each with tolerable risk shows how much the existing measures achieve and what is still missing.

Risk register. Record each scenario with its ratings, owner, chosen treatment and status, so decisions can be tracked and revisited.

Exam angle: tolerable risk is the asset owner's decision; use one set of scales across safety, business and cyber risk; unmitigated risk ignores existing countermeasures, residual risk includes them.

6 · Initial risk assessment and partitioning (ZCR 1–4)

The 62443-3-2 workflow: seven zone and conduit requirements and one decision

ZCR 1, identify the SUC. Set the boundary and list the access points. The SUC can be a whole site's IACS or one process unit, and it should include the supporting systems that can affect it, such as historians, engineering workstations and domain services.

ZCR 2, initial risk assessment. Ask what the worst credible outcome would be if the SUC, or a major part of it, were compromised, before counting any countermeasures. PHA results and business impact data feed this step. The answer shows how critical each part is and guides the partitioning.

ZCR 3, partition into zones and conduits. Its sub-requirements say what to keep apart:

ZCRRequirementWhy
3.1Establish zones and conduitsGroup assets by common security requirements
3.2Separate business and IACS assetsBusiness systems face different threats and change faster
3.3Separate safety-related assetsProtect the last automated layer of protection
3.4Separate temporarily connected devicesLaptops and portable media come and go with unknown hygiene
3.5Separate wireless devicesRadio extends the network beyond physical walls
3.6Separate devices connected through external networksRemote and cloud connections cross trust boundaries

Partitioning tips. Every asset in a zone inherits the zone's target security level, so a zone that mixes critical and minor assets must either meet the highest need or be split. Group by criticality, function, location, ownership or technology. Too many tiny zones are hard to manage; one big zone lets a single compromise spread. Keep conduits few, with known and documented flows.

ZCR 4, compare initial risk with tolerable risk. Where the initial risk is above the tolerable level, carry out a detailed risk assessment of the zones and conduits (ZCR 5). Where it is not, go straight to documenting the requirements (ZCR 6).

Exam angle: know the order of the steps and the six ZCR 3 sub-requirements. Safety-related assets, wireless, temporary devices and external connections each get separated.

7 · Detailed risk assessment (ZCR 5)

The detailed assessment runs for each zone and conduit whose initial risk was above the tolerable level. It has thirteen steps:

ZCRStepWhat you do
5.1Identify threatsThreat sources, their capability, the vectors they could use and the assets affected
5.2Identify vulnerabilitiesWeaknesses in assets, architecture, configuration and procedures, drawn from the vulnerability assessment
5.3Determine consequence and impactWhat happens if each threat succeeds, rated on the company's consequence scale with PHA input
5.4Determine unmitigated likelihoodHow likely each threat is to succeed, ignoring existing countermeasures
5.5Determine unmitigated cyber riskCombine likelihood and consequence on the risk matrix
5.6Determine SL-TThe target security level, per foundational requirement, that the zone or conduit needs
5.7Compare unmitigated risk with tolerable riskDecide where countermeasures are needed
5.8Identify and evaluate existing countermeasuresWhat is already in place, and how well it actually works
5.9Reevaluate likelihood and impactRate them again with the existing countermeasures counted
5.10Determine residual riskThe risk that remains with existing countermeasures
5.11Compare residual risk with tolerable riskIs it acceptable now?
5.12Identify additional countermeasuresTechnical, procedural or physical measures until residual risk is tolerable
5.13Document and communicate resultsRecord the results and report them to the people who need them

Steps 5.9 to 5.12 form a loop: add countermeasures, re-rate, and compare again until residual risk is tolerable or the owner formally accepts what remains.

Setting SL-T (ZCR 5.6). The owner decides how risk maps to a target level, for example from the capability of the credible threat using the SL definitions, or from a table that maps unmitigated risk to a level. The result is a vector, so think about which foundational requirements matter for this zone: integrity and availability usually dominate in a control zone, while confidentiality may matter more for a server holding recipes or credentials.

Crediting existing countermeasures (ZCR 5.8). Credit only what works as installed. A firewall with "any to any" rules, antivirus with signatures a year old, or a procedure nobody follows reduces nothing. Verify by configuration review or observation rather than by asking.

Running the workshops

Exam angle: unmitigated likelihood and risk come before existing countermeasures are counted; SL-T is set inside ZCR 5 (5.6); residual risk is compared with tolerable risk, and countermeasures are added until it is tolerable.

Risk assessors must be able to read process safety studies, because they hold the consequences that make IACS risk different.

TermMeaning
PHAProcess hazard analysis: a systematic review of a process's hazards and safeguards
HAZOPThe most common PHA method: guide words (no, more, less, reverse, other than) applied to parameters such as flow, pressure, temperature and level at each node reveal deviations, their causes, consequences and safeguards
LOPALayer of protection analysis: checks whether independent protection layers reduce the risk of a scenario enough, and sets SIL targets for safety instrumented functions
IPLIndependent protection layer, such as a SIS function, a relief valve or a dike
SIF and SILA safety instrumented function and its safety integrity level

What a cyber assessment takes from the PHA

Cyber PHA (also called a cyber HAZOP) adds a cyberattack as a cause. For each scenario where a programmable system is a cause or a safeguard, the team asks whether an attacker could produce the deviation and defeat the programmable safeguards at the same time. Where the answer is yes, it recommends safeguards an attack cannot reach, or stronger security countermeasures.

PHA entryCyber question
High pressure if the control valve fails open; safeguards are a SIS high-pressure trip and a relief valveCould an attacker drive the valve open through the control system and also bypass the trip? The relief valve still limits the outcome, so the worst case is a release to flare, not a rupture

Securing the safety instrumented system

Exam angle: the PHA supplies consequences; safeguards an attack cannot reach limit the worst case; one attack defeating both cause and safeguard is the key cyber PHA question; the SIS gets its own zone.

9 · Countermeasures, SL-T and residual risk

Kinds of countermeasure

By formExamples
TechnicalFirewalls, authentication, allowlisting, encryption, monitoring
ProceduralPolicies, management of change, training, account reviews, backup procedures
PhysicalLocked rooms and cabinets, port blockers, controller key switches

They can also be described by effect: preventive (stop the attack), detective (notice it) and corrective (recover from it). A good set covers all three.

Weighing candidates. ISA's course objectives call for evaluating cost, complexity and effectiveness:

FactorQuestions to ask
EffectivenessWhich scenarios does it address, and by how much does it cut likelihood or consequence?
CostPurchase plus the lifecycle cost: licenses, maintenance, staff time
ComplexitySkills needed to run it; operational burden; chance of misconfiguration
Operational impactLatency, availability, safety, vendor support and warranty
FeasibilityWorks with legacy equipment? Needs an outage? How long to implement?

Prefer measures that cover many scenarios at once, such as segmentation; layer them for defense in depth; and never pick one that compromises an essential function.

A simple cost-benefit check. If a countermeasure lowers annualized loss expectancy from $100,000 to $20,000 and costs $30,000 a year to own, it returns $50,000 a year. Few cyber risks have reliable numbers, so most teams make the same comparison qualitatively.

From SL-T to design. The SL-T vector from ZCR 5.6 decides which 62443-3-3 system requirements and requirement enhancements apply to each zone and conduit. Components and system designs should have an SL-C that meets the SL-T for every foundational requirement. Where one falls short, add compensating countermeasures at the zone level, or document the gap as residual risk. The SL-A is verified after implementation.

Gaps in policies and procedures are countermeasures too. Missing account management, untested backups or no management of change often show up when practice is compared with 62443-2-1 or the owner's own policies. Their fixes are procedural and usually cheap.

Residual risk. Record the residual risk for each zone and conduit. If it is still above the tolerable level and no reasonable countermeasure exists, the options are to redesign (avoid the risk), transfer part of it, or accept it formally at the right level of management with a written justification.

Presenting recommendations. Rank them by risk reduction. Separate quick wins, such as changing default passwords or removing unused remote access, from longer projects such as re-architecting the network, and lay them out as a roadmap.

Exam angle: weigh effectiveness, cost and complexity; credit only countermeasures that actually work; SL-C must meet SL-T or be compensated; residual risk is accepted by the asset owner.

10 · The CRS, reports and approval (ZCR 6–7)

ZCR 6 requires the cybersecurity requirements, assumptions and constraints to be written down in a cybersecurity requirements specification (CRS). Its contents follow ZCR 6.1 to 6.9:

ZCRContentWhat it holds
6.1Cybersecurity requirements specificationThe document itself, assembled from the items below
6.2SUC descriptionName, purpose, boundary and a high-level description of the system
6.3Zone and conduit drawingsThe partitioned architecture
6.4Zone and conduit characteristicsFor each zone and conduit, the attributes listed below
6.5Operating environment assumptionsFor example the physical setting, staffing and support arrangements
6.6Threat environmentThe threat sources and vectors that were considered
6.7Organizational security policiesThe owner's policies the design must follow
6.8Tolerable riskThe owner's risk criteria
6.9Regulatory requirementsLaws, regulations and standards that apply

Zone and conduit characteristics (ZCR 6.4) include a name or identifier; the accountable organization; the logical boundary and, where relevant, the physical boundary; the safety designation; the logical and physical access points and the data flows through them; the connected zones and conduits; the assets with their classification, criticality and business value; the SL-T; the applicable security requirements and policies; and the assumptions and external dependencies.

Who uses the CRS. Integrators and suppliers design to it, procurement buys against it, factory and site acceptance tests verify against it, and maintenance keeps it current through management of change. It is the hand-off from the assessment phase to the design phase.

ZCR 7, asset owner approval. Management with the authority to do so reviews and signs off the CRS, the residual risk and the assumptions behind them.

The assessment report

SectionContentsMain reader
Executive summaryOverall posture, top risks, key recommendationsManagement
Scope and methodScope, team, dates, rules of engagement, standards used, limitationsEveryone
FindingsVulnerabilities and gaps with evidence, mapped to requirementsEngineering and security
Risk registerScenarios with unmitigated and residual ratingsRisk owners
RecommendationsPrioritized countermeasures with effort and cost, as a roadmapManagement and engineering
AppendicesInventory, network and zone and conduit diagrams, raw dataEngineering

The report is a map of the site's weaknesses: classify it as confidential, limit distribution and protect stored copies.

Documents to keep current after the assessment: the asset inventory, network and zone and conduit diagrams, the CRS, the risk register, and the policies and procedures. Management of change and periodic reassessment keep them accurate.

Exam angle: the CRS is the output of ZCR 6 and the input to design; it holds the SL-T of each zone and conduit, the assumptions, the threat environment, tolerable risk and regulations; the asset owner approves it in ZCR 7.

Exam traps, tie-breakers and sources

Pairs that get mixed up

Easily confusedThe difference
Initial and detailed risk assessmentZCR 2: worst-case consequences for the whole SUC, before partitioning. ZCR 5: threats, vulnerabilities and likelihood per zone and conduit
Unmitigated and residual riskBefore existing countermeasures are counted; after they are counted
SL-T, SL-C, SL-ASet in ZCR 5.6 for zones and conduits; offered by products; verified after implementation
Threat and vulnerabilityWho or what could cause harm; the weakness they would use
Gap, vulnerability and risk assessmentCompare with a standard; find weaknesses; combine threat, weakness and consequence
Vulnerability scan and penetration testFinds likely weaknesses; proves impact by exploiting them
CVSS score and riskTechnical severity of a flaw; what it means for your plant
CRS and assessment reportRequirements handed to design; findings and recommendations for the owner
ZCR 3.3, 3.4, 3.5, 3.6Separate safety-related assets; temporarily connected devices; wireless; external network connections

First steps the exam likes to ask about

Tie-breakers for BEST answers

  1. Passive methods before active ones on a running system.
  2. Credible scenarios over the worst imaginable one.
  3. The asset owner sets tolerable risk and accepts residual risk.
  4. Credit only countermeasures that are verified to work.
  5. Write assumptions down; they belong in the CRS.
  6. Choose the cross-functional answer over one person or one tool.

Sources

These notes paraphrase and summarize. They do not reproduce the standards, so check details against your licensed copies of 62443-3-2 and IEC 61511.