We are getting better every week, every month at measuring the things we can see.
Equipment ‘tells’ us when it is degrading. Systems tell us when a process has been missed. Audits tell us whether we have complied. Incident reports tell us what went wrong. Alas, some of the most important risks in our terminals don't announce themselves quite so clearly.
Fatigue. Behavioural drift. The small shortcuts, the near misses, the things that become normal because nothing has gone wrong yet.
I have always believed that safety is as much about what we don't see, as what we do. The challenge, of course, is how do you manage a risk that your systems can’t spot?
This is where I think AI will have a genuinely interesting role to play. It will not, and should not, replace the experience and judgement of safety leaders; but it can assuredly give them a much better view of signals that are otherwise easy to miss. Perhaps, more importantly, it can help us move from asking what went wrong, to asking what might go wrong next.
This is the theme running through this week's newsletter. I’d very much like to hear from you- so do comment or write in. Better still, let’s connect.
I will be at CTAC in Bangkok next week, looking forward to sharing the stage with some leading minds in our field. Maybe see you then!
Arjun
Fatalities from fatigue.
Fatalities from equipment failure.
Both are real. Yet while we have rigorous systems to detect when machines are degrading, fatigue is often reduced to simply, “human error.”
We must change this.
Fatigue directly affects reaction time, judgment, risk perception. It can build gradually across shifts, often without triggering the same alerts or controls we have for equipment-related risks.
The question is not whether fatigue is a risk, it assuredly is. The question is whether we are measuring and managing it like one.
Because some of the most important safety risks are the ones traditional systems don't see.
[👉🏼 You can also watch my take on YouTube if you prefer.]
Safety audits, compliance checks and incident reports are essential. But they often tell only part of the story.
Some of the greatest risks remain hidden in spreadsheets, ignored near-misses, normalized unsafe habits and the shortcuts created by inefficient processes.
These are the silent risks—invisible to traditional reporting, yet capable of becoming the next major incident. The challenge for safety leaders is to move beyond simply tracking completion and start looking at engagement, behaviour, trends and early warning signals.
This is where AI can come in, make a difference using real-time data to identify patterns and surface risks before they become problems.
The objective is not more reporting. The aim to build greater situational awareness– seeing the behavioural drift and subtle cues that traditional systems overlook.
👉🏼 Read more on LinkedIn
Yet, the reality is that some risks are still invisible to us. Which bring us to the most interesting question: can we predict them before they become incidents?
The world’s leading terminals are moving beyond reacting to incidents. They are asking a different question:
“How can we see risk coming?”
At the most advanced terminals, AI and real-time data are being used to predict equipment failures, identify high-risk operational windows and uncover behavioural patterns that can lead to incidents.
The shift is significant—from responding to anticipating. As are the benefits, which extend beyond safety: better equipment utilisation, stronger operational continuity, faster decision-making and greater confidence across the organisation.
The competitive gap is widening. While some terminals are still digitising checklists and reports, others are already using predictive intelligence to identify risks before they become problems.
This brings the conversation full circle.
Fatigue showed us the risks we struggle to measure. Silent risks showed us what traditional systems fail to see. Predictive safety shows us what becomes possible when we start looking ahead.
We have all become good at collecting data about our terminals- let’s become proactive with turning data into foresight… and foresight into prevention.
👉🏼 Read more on LinkedIn
