The new reality of Operational Safety: why traditional models are no longer enough

For decades, operational safety was approached through a relatively stable framework: clear rules, documented procedures, periodic training, and on-site supervision.That model — with variations — helped reduce accidents and professionalise risk management across many industries.However, operational contexts have evolved faster than safety models.

Today, many organisations face an uncomfortable paradox: more rules, more procedures, and more training, yet results are not improving at the same pace.The cause is rarely a lack of intent, or even a lack of investment. It lies in a growing gap between how safety is designed and how work is actually carried out.

The core issue: safety is still designed “from the desk”

In practice, most current safety systems continue to rely on three assumptions that no longer hold:

1. That people remember and correctly apply procedures at the critical moment

2. That risk is mainly controlled before the task begins (inductions, permits, briefings)

3. That the main problem is attitude or behaviour (when the real issue is the difficulty of executing consistently)

Operational reality shows something different.

Today, most incidents — and nearly all serious accidents — occur during execution, not because rules are missing, but due to deviations in how work is performed: shortcuts, omissions, partial interpretations, or decisions made with incomplete information.Not because people want to do things wrong, but because the system fails to support them when they need it most.

Increasingly complex processes, with less cognitive margin

Modern operations are objectively more complex:

• More procedures per task

• More compliance requirements

• More interaction between departments, contractors and systems

• More pressure for operational continuity and efficiency

At the same time, the workforce profile has changed:

• High turnover and shorter tenure

• Difficulty in recruiting and retaining highly trained teams

• Lower familiarity with specific processes

• Less effective training time

The outcome is predictable: perfect execution is expected in environments where remembering and applying everything correctly is unrealistic.

Expecting someone to recall dozens of critical steps, safety conditions, permits, isolations, PPE and validations without real-time support is not a safety strategy. It is a gamble.

The gap between prescribed work and real work

This is not new in industrial safety — but it is more visible than ever.

• Prescribed work lives in manuals, PDFs, folders, intranets and procedures

• Real work happens on site, with noise, interruptions, uncertainty and time pressure

The wider this gap becomes, the higher the risk.

Most organisations still try to close it with:

• More documentation

• More pre- and post-controls

• More audits

• More general training

But the problem is not the volume of information. It is about timing and context.

The critical moment of safety is execution

When you observe where safety systems fail, the same pattern appears repeatedly:

• The procedure existed, but was not applied correctly

• The permit was approved, but conditions changed

• PPE was defined, but not used

• The risk was known, but not detected in time

All of this happens while work is in progress.

That is precisely where most traditional tools stop operating. And that is where real safety is now determined.

Why safety must shift from “Controlling” to “Supporting”

Historically, safety has been preventive and reactive:

• Prevent beforehand

• Investigate afterwards

What is missing is assistance during execution.

Organisations need systems that:

• Guide tasks step by stepm

• Prevent critical steps from being forgotten

• Adapt safety to the real context of work

• Detect deviations in real time, not weeks later

• Turn rules into concrete action

This is not about replacing procedures or responsibilities. It is about bringing safety to the point where risk is actually created.

The new category of Solutions: Safety integrated into operations

In this context, a new type of solution is emerging — not positioned merely as “safety systems”, but as operational platforms focused on safe execution.

Systems that:

• Integrate permits, tasks, risks, controls and knowledge into a single workflow

• Support workers throughout the task

• Make explicit what previously depended on memory

• Provide real-time visibility for supervision and prevention

• Transform operational data into organisational learning

These solutions recognise that safety is not an event, but a dynamic process that unfolds as work progresses.

Conclusion: The Safety of the future is defined in execution

Operational safety can no longer rely solely on:

• More rules

• More paperwork

• More pre- or post-controls

The real challenge today is not defining what should be done — but ensuring it is done correctly, at the right moment, under real conditions.

Organisations that understand this will stop chasing formal compliance and start building genuinely safer operations.

Because in practice, safety does not fail in design. It fails when it does not support execution.

Safety cannot depend on human memory at the critical moment. Either execution is supported — or risk is accepted.