Risk Ranking, Severity, Frequency, and Corrective Action

Risk Ranking, Severity, Frequency, and Corrective Action

A construction site generates more hazards than any one person can fix at once. Risk ranking lets the CHST and the project team decide what needs immediate action, what can be scheduled, and what only needs monitoring. The goal is not a perfect mathematical score; it is a consistent, defensible decision that protects workers from the most serious and most likely harm first.

Core risk factors

Most simple systems combine severity (how bad the credible outcome could be) with likelihood or frequency (how often or how long workers are exposed). Complete field judgment also weighs how many workers are exposed, whether the hazard is detectable, whether current controls are reliable, and how fast conditions can worsen. A shallow, empty trench is low immediate exposure; the same trench with workers inside and spoil piled at the lip becomes an urgent cave-in hazard — soil weighs roughly 3,000 pounds per cubic yard, so a partial collapse can kill before rescue arrives.

Using a simple risk matrix

A risk matrix helps organize judgment. Severity may be scored 1–5 and likelihood 1–5; the product gives a priority score (1–25). This is only useful when the ratings come from credible field information. The matrix must never hide common sense: a permit-required confined-space atmosphere, an unprotected six-story edge, or live electrical work can demand an immediate stop or escalation no matter what number the matrix produces. A practical construction ranking ties the score to action:

• Imminent danger — stop work or remove workers until the hazard is controlled.
• High — correct before the task continues or before exposure begins.
• Moderate — assign corrective action with a near-term deadline plus interim protection.
• Low — correct as scheduled, monitor for change, prevent recurrence.

Ranking must be linked to authority. If the CHST identifies imminent danger but lacks direct authority over a subcontractor, the issue is escalated immediately through the project chain of command, and documentation records the condition, location, people notified, interim controls, final correction, and verification.

Corrective-action quality

A corrective action states what will be done, who owns it, when it is due, and how completion will be verified. "Replace bad ladder" is weak. "Remove the damaged ladder from service, tag it Do Not Use, notify the sub's supervisor, and verify a compliant replacement before access resumes" prevents continued exposure and creates a checkable endpoint. Corrective actions should also reach the root or contributing cause when a hazard recurs.

If crews keep removing guardrails to load material, reinstalling the rail after each observation will not hold; the project may need a designated loading bay, a removable-guardrail procedure with anchor points, supervision, and schedule coordination. If silica controls keep failing because water is unavailable, the fix is water-supply planning, hose routing, equipment maintenance, and a supervisor check before any cutting starts.

Prioritizing without ignoring lower risks

High-severity hazards get prompt attention, but lower risks still need tracking. Poor housekeeping, missing secondary-container labels, blocked egress, or minor tool defects can both cause injuries and signal weak control discipline. The difference is timing and escalation, not whether the item closes. A missing guardrail at height and a missing label do not get the same response, but both get a closure date and owner.

Risk ranking also sharpens communication: instead of "this is unsafe," the CHST can say, "this is high severity because a worker could fall two stories, immediate because the crew is within six feet of the edge now, and unreliable because caution tape cannot arrest a fall." That framing is hard for operations to dismiss.

Verification and follow-up

A corrective action is complete only when the hazard is controlled and the control is verified — usually by field observation, not just a photo. For recurring hazards, trend review shows whether corrections last. If the same hazard reappears across multiple inspections or multiple subs, treat it as a system issue rooted in planning, supervision, training, procurement, or schedule pressure.

A worked ranking example

Suppose a walk turns up four conditions on the same morning: an unprotected leading edge at the eighth floor with a crew working within four feet of it; a partially full chemical drum missing a secondary label in a locked storeroom; a frayed extension cord with the ground pin intact; and a slip hazard from tracked-in mud at a single entrance. Rank by severity times exposure. The leading edge scores high-severity (fatal fall) and immediate exposure — it is imminent danger, so stop or remove the crew now. The frayed cord is moderate severity (shock) but easily corrected, so tag it out of service today.

The mud is low severity but real, so squeegee and mat it on a same-day schedule. The unlabeled, locked drum is lowest priority — no one is exposed — yet it still gets an owner and date. Notice that ease of fixing the mud must not pull attention away from the edge, which is the most common ranking mistake.

Common exam traps

Watch for items that reward fixing the easiest or most visible problem first; the correct answer addresses the highest combination of severity and exposure even when it is harder or politically awkward. Another trap presents a low matrix score for a clearly lethal condition — the right response treats the imminent danger as a stop-work regardless of the number. For the exam, remember the matrix is a prioritization tool, never a substitute for compliance or professional judgment; some conditions require immediate control regardless of paperwork.