A GTC 45 hazard matrix: a practical example and criteria for managing it in software
A hazard matrix adds value when it explains what happens at work and supports decisions about controls. Moving it into software requires preserving that logic: task, hazard, exposed people, assessment, measures and review. Digitising a table can make information easier to retrieve while still leaving the management of risk unfinished.
In Colombia, article 2.2.4.6.15 of Decree 1072 of 2015 requires a systematic methodology for identifying hazards and assessing risks. It does not establish GTC 45 as the only mandatory method. Technical selection and application should reflect the activity and hazards, including supplementary assessments where appropriate.
Define the methodology before configuring the matrix
If your organisation uses GTC 45, identify the edition and rules adopted by its competent team. Keep the definitions and criteria required for consistent application available. Do not mix scales from different editions or methodologies without a documented technical reason and an explanation that reviewers can follow.
As an example of institutional application, the procedure of Hospital San Agustín de Fonseca identifies GTC 45 as its chosen tool. This demonstrates adoption by one organisation; it is not a universal template that should be copied into every business regardless of context.
The Explora entry on risk matrices explains the instrument’s purpose. In the software project, ask how the chosen method is retained and who can change its configuration. A method name on a screen is insufficient if its underlying settings cannot be explained.
An illustrative example: receiving goods
Imagine a warehouse where someone checks incoming goods near a forklift route. A useful description begins with the task, location, exposure circumstances and affected people. “Accident” describes a possible outcome; it does not replace identification of the hazard or explanation of how the interaction occurs.
The team checks existing controls in the workplace: physical separation, routes, visibility, unloading arrangements and instructions. Distinguish controls that operate from those merely planned or still requiring verification. Participation by the people doing the work allows the description to be compared with actual operating conditions.
For this fictional scenario, the descriptive part of a row could be organised as follows. Controls are separated by source, environment and person, as in the institutional procedure cited above:
| Field | Qualitative example |
|---|---|
| Process, area and task | Receiving, unloading bay: checking labels as goods arrive. A routine activity within normal operations. |
| Hazard and possible effects | Interaction between a moving forklift and a pedestrian at a shared crossing; potential impact or run-over injury. |
| Exposed people | Receiving staff, the operator and other people using the crossing. Verify who accesses it and when. |
| Control at source | A pre-use equipment check is specified; verify that it takes place and how identified defects are resolved. |
| Control in the environment | A marked route, with no physical separation at the crossing described. Observe traffic during an unloading operation. |
| Control relating to the person | Traffic instructions have been provided. Check understanding and application with the people performing the task. |
| Proposed improvement and follow-up | Operations and OSH assess separation of the routes. Record the decision, owner, date and verification in use before treating the measure as effective. |
The row distinguishes observations, controls requiring verification and proposed measures. Risk assessment and acceptability still need to be determined using the adopted method; this is neither a complete matrix nor an assessment of a real facility.
Competent personnel assess the risk using the adopted criteria. This example deliberately assigns no scores: without knowing conditions, frequency, controls and consequences, a numerical answer would create false precision. The Explora entry on risk assessment explains the technical judgement behind the matrix.
Connect the result to a verifiable control
Where improvement is required, record the measure, owner, resources, deadline and verification criterion. A traffic-related intervention may involve design, installation, communication and checking performance in use. These activities are connected but do not necessarily belong to the same person or department.
Distinguish the current assessment from an estimate of the expected result after implementation. Risk should not appear reduced solely because an action has been scheduled. Once implemented, the team should check how the control works and review the assessment as appropriate, preserving the reason and evidence for the change.
What to test in the software demonstration
- Structure: connect processes, tasks, sites, hazards and exposed people.
- Criteria: identify the method, version and technical review owners.
- Traceability: preserve assessment changes and the reasoning behind them.
- Actions: link measures to owners, deadlines and verification.
- Consultation: retrieve the current position and relevant previous versions.
Add a test in which the vehicle route changes. Check how the affected assessment is identified and review begins. The decree provides for updates at least annually and following the events and changes it specifies. A calendar reminder alone does not cover every such situation.
Keep the matrix connected to planning
A fully completed matrix can still contain outstanding controls. Review progress, overdue measures and resource decisions needed. The Explora entry on preventive planning provides general guidance for maintaining the connection between assessment results and the actions intended to address them.
Include the work in the SG-SST annual plan and use these scenarios when choosing software in Colombia. When considering Sabentis OSH software, request a demonstration connecting assessment, action and verification using your organisation’s methodology and a clearly defined scope.
Checklist for validating a migrated hazard matrix
Before accepting migration, select contrasting tasks and compare the original information with what the team will use. The review should identify changes in meaning as well as empty cells or missing attachments.
- Select a routine task, an occasional intervention and a situation involving several exposed people; verify that each retains its context.
- Find controls written in the future tense and check that they are not treated as existing protection in the current assessment.
- Compare a row with dated observations or photographs, avoiding the use of old images as proof of present conditions.
- Check how an unknown input is recorded: it should require verification rather than automatically become an absence of hazard.
- Ask another assessor to reconstruct a decision using the recorded evidence and method, then identify any missing information.
Frequently asked questions
Is GTC 45 compulsory for every Colombian company?
Decree 1072 requires a systematic methodology but does not prescribe GTC 45 as the sole method. If adopting this guide, document its application and check for requirements specific to the hazards being assessed. A template name does not, by itself, establish an adequate risk assessment.
Can Excel scores be imported without checking the formulas?
First compare a sample against the original spreadsheet and the adopted technical criteria. Decimal separators, empty fields or different scales can change results. Retain original scores and explain discrepancies before approving the calculation that the team will use for its operational decisions.
Does a general matrix replace exposure measurements?
It should not be used to omit a necessary specific assessment. A row can identify noise, chemicals or other hazards and link to the relevant study. Connect its findings and actions to the matrix while preserving the method, date and conditions under which the assessment took place.
What if two assessors assign different priorities?
First check whether they observed the same task, exposure and controls and applied the same criteria. Document the reasons for their difference and resolve it with competent personnel. Averaging scores without investigating the discrepancy may conceal a significant condition or an input that remains uncertain.
Legislation and consulted sources
- Decree 1072 of 2015, article 2.2.4.6.15. Colombia: scope of hazard identification and risk assessment. Spanish text.
- GTC 45, second update, 2012. ICONTEC guide consulted in Colombia’s General Accounting Office official repository; a methodological reference, not a universal legal obligation. Spanish text.
Sources consulted on 6 October 2026.
