IoT Solutions
Transportation Hubs
Every terminal has a near-miss form. An internal truck turns out of a stack row too fast. A straddle carrier passes close to a lasher. Nobody is hurt, someone fills in the form, and the report joins a folder.
Safety teams already know these reports are valuable. A near-miss is the same event as an accident, minus the injury, and it is the cheapest warning a terminal will ever get. The trouble is rarely that near-misses go unreported. The trouble is that a report records that something almost happened, and almost never where, when and why.
This article looks at what a near-miss report usually leaves out, how a replay and a heat map fill the gap, and what a safety team can change once it sees the pattern.
What is a near-miss, and why does it matter in a terminal?
A near-miss is an event that caused no harm but could have. The Health and Safety Executive (HSE) defines it in its investigation workbook, HSG245, as “an event that, while not causing harm, has the potential to cause injury or ill health.”
The same workbook makes the point that matters most for a yard: “It is often pure luck that determines whether an undesired circumstance translates into a near miss or accident.” The truck and the lasher in the opening example were one second apart from a very different report.
In ports, that luck is tested by vehicles more than anything else. The Occupational Safety and Health Administration (OSHA) built its Longshoring and Marine Terminals: Fatal Facts from 42 actual OSHA case file summaries of longshoring deaths. Its finding is blunt: the most frequent cause was workers being struck by or run over by vehicles such as trucks, front-end loaders or forklifts.
The HSE’s own page on workplace transport in ports and docks lists the same hazards a terminal manager sees every shift: reversing vehicles, unsegregated vehicle and pedestrian access, and vehicles with limited visibility, including straddle carriers and reach stackers. So near-misses between machines and people are not a side topic. They are the early signal for the most serious kind of event in the yard.
What does a typical near-miss report leave out?
A typical report is written after the shift, from memory, and it rarely says where the paths crossed or why. It says “near the reefer stack” or “lane 4, afternoon”, and the details that would point to a fix are gone.
Most forms cannot capture exactly where the two paths crossed, which direction each person or machine was moving, or whether the same thing happened yesterday at the same spot.
HSG245 asks investigators to “describe the chain of events leading up to, and immediately after, the adverse event” and to note positions before and after. A two-line form written at the end of the shift cannot do that. Without the chain of events, the usual fix is a toolbox talk and a reminder to “stay alert”. The crossing stays where it was.
| What you need to know | Paper or form report | Replay on the terminal map |
|---|---|---|
| Where the paths crossed | “Near block 2” | The exact place on the terminal map |
| How the machine was moving | Witness memory | Its route and direction before the event |
| What else was going on | Rarely recorded | The alert timeline, plus related alerts at the same place, machine or badge |
| Whether it happened before | Someone has to remember | Same place over 30 days, on the heat map and in related alerts |
| When the review starts | Days later | The same shift |
How does a replay change a near-miss investigation?
A replay turns a near-miss from a sentence into a moment you can watch again on the terminal map. Everyone in the review starts from the same picture of where the truck was and how it was moving.
In Mapsted Yard Watch, a near-miss alert lands in one inbox together with every other alert. From there, View replay opens that truck’s path for the shift, at the minute of the alert. The camera can follow the truck, so the minute before the event is visible, not reconstructed.
Timing matters as much as detail. HSG245 says adverse events “should be investigated and analysed as soon as possible”, because “memory is best and motivation greatest immediately after an adverse event.” A replay lets the review start in the same shift, while the people involved still remember why they did what they did.
There is a second benefit. The HSE team and the operations team look at the same replay, so the conversation starts from what happened, not from two versions of it. That shortens the meeting and keeps it about the layout rather than about blame.
What does a near-miss look like, step by step?
Take a common case: an internal truck brakes hard at the end of a stack row because a lasher is crossing on foot towards the quay. Nobody is hurt.
With a form, the lasher’s supervisor writes two lines at the end of the shift. With a replay, the HSE lead opens the alert in the same shift and watches the truck’s path in the minute before it.
The replay shows something the form never would: the truck left its lane to cut through the end of the row. Asked the same shift, the lasher says that is the usual walking route to the quay, and the truck waits at that crane for that shift show why the driver took the short cut. Both were doing what made sense to them at the time.
That is a different conversation from “stay alert”. It points at two things the terminal can change: the queue that pushes trucks onto the short cut, and a walking route that crosses a truck lane. It also takes minutes, not a week of interviews.
How do you turn single near-misses into a pattern?
You count them by place over weeks, then look at where they cluster. One near-miss is an event; thirty at the same corner is a design problem.
HSG245 supports this directly: “if you have had a number of similar adverse events, it may be worth investigating, even if each single event is not worth investigating in isolation.” The difficulty is seeing that the events are similar when each one arrives as a separate form.
In Yard Watch, the heat map shows where near-misses cluster across the yard, and the alerts inbox ranks the top 5 places for near-misses over the last 30 days. The pattern is usually obvious once it is drawn: a pedestrian route that cuts across a truck lane, a blind corner at the end of a stack, or a gate where trucks and crews arrive at the same time every shift change.
The OSHA near-miss reporting policy template puts it simply: “Near misses are symptoms of undiscovered safety concerns.” A heat map is how those symptoms become a diagnosis.
What can a safety team change once it sees the cluster?
It can change the layout, then check the heat map again to see whether the cluster went away. That loop of change and re-check is what turns near-miss data into a safer yard.
The HSE’s guidance on separating pedestrians and vehicles gives a useful menu. It says “the most effective way” is to make routes entirely separate, that pedestrian routes should follow “desire lines” so people stay on them, and that crossing points should be marked so everyone can see clearly in all directions. It also suggests limiting vehicles on a roadway “during a shift changeover, when many pedestrians are likely to be crossing.”
| Cluster you see on the heat map | Change to consider |
|---|---|
| Walking route across a truck lane | Move or fence the route along the desire line |
| Blind corner at a stack end | Marked crossing point, mirror or stop line |
| Gate busy at shift change | Stagger arrivals or hold vehicles at changeover |
| Short cut around a crane queue | Fix the queue, then close the short cut |
| Reversing near the quay edge | One-way lane or a turning area |
After the change, look at the same area again over the next 7 or 30 days. If the cluster has moved one row along, the fix moved the problem rather than removing it. If it has gone, you have a before-and-after story that a form folder could never tell.
Why should operations and finance care about near-misses?
Because near-misses and slow operations often share a cause. A crossing where trucks brake for pedestrians is also a crossing where trucks queue, so fixing the layout tends to make the yard both safer and faster.
That is why near-miss data deserves the attention of the COO, not only the HSE team. In Yard Watch, Insights turn the last 7 or 30 days into plain “What to do” cards for operations, maintenance, health and safety, and site access, each built from that area’s own numbers.
It also matters for the conversation with your insurer and your auditors. “We had thirty near-misses at this crossing, we changed the route, and here is the month after” is a much stronger account than a folder of forms. Yard Watch builds incident timelines, an insurer report covering the last 30 days with CSV export, and an audit trail of who did what, and when, automatically, so a terminal can run a safer, faster yard and prove it to its insurer and auditors.
There is a quieter benefit too. The OSHA template asks that the steps taken after a near-miss investigation be reported to management and to all staff. When crews see that a report leads to a visible change in the yard, such as a new crossing or a moved route, they report more. More reports mean a clearer picture, and the cheapest safety data you have gets better every month.
What are the limits of near-miss data?
Near-miss data shows where risk concentrates; it does not remove the risk by itself. The value comes from the changes a terminal makes and from checking whether they worked.
A heat map can also mislead if it is read alone. A busy area will show more events simply because more people and machines pass through it, so compare clusters with the traffic around them before you redesign a block. And a quiet area is not automatically safe. It may simply be an area where people do not yet report.
None of this replaces your procedures. The HSE and OSHA guidance cited here still applies in full; a replay and a heat map make it easier to follow, not optional.
What should you look for in a near-miss tool?
Look for five things: replay, one inbox, patterns over time, clear answers on who sees names, and named functions that keep working when the network does not. Each one closes a gap that a paper form leaves open.
-
Can we replay the shift on the map? A replay shows the chain of events before and after the alert, not a summary written from memory.
-
Do all alerts land in one place? Near-misses, SOS calls, access alerts and machine alerts in separate systems mean nobody sees the whole picture.
-
Can we see patterns over 7 or 30 days? A heat map or a ranked list of hot spots turns single events into a design decision.
-
Which screens show names, and which person views are logged? Look for per-crew time reports, and ask the supplier to show you each screen that names a worker and each look-up of one person that needs a stated reason and is written to a log.
-
What keeps working when the network does not? Yards lose cloud and cellular connections, so alarms, access rules and SOS relay should keep working on site when that happens.
Privacy deserves extra weight. Crews will only trust a system that protects them rather than watches them, and a screen-by-screen answer on who sees names is the simplest test of whether a supplier has thought about it.
Mapsted Yard Watch shows every truck, machine and crew member on your terminal map live, replays any one of them for a shift or time window, and gives you one inbox for every alert and a heat map of where near-misses cluster. It is already used to track 1,000+ assets across multiple sites. Book a private demo to see a near-miss go from alert to replay to pattern.
Mapsted Yard Watch supports, and does not replace, your safety procedures and emergency services.
Frequently asked questions
What counts as a near-miss in a container terminal?
A near-miss is any event that caused no harm but had the potential to cause injury or ill health. In a terminal, typical examples are a truck braking hard for a person on foot, a straddle carrier passing too close to a lasher, or a vehicle reversing into a lane where someone is walking. The HSE uses this definition in its HSG245 investigation workbook.
Why are near-misses worth investigating if nobody was hurt?
Because luck, not good design, often decides whether an event becomes a near-miss or an accident. A near-miss shows the same weakness as an accident without the injury, the legal pressure or the damaged trust. Investigating it is easier, the people involved speak more freely, and the fix you make protects against the more serious version of the same event.
How does a heat map help a terminal safety team?
A heat map shows where near-misses cluster across the yard over a chosen period, such as the last 30 days. Single reports rarely look connected, but drawn on the terminal map they often point at one crossing, one stack end or one gate. That turns a pile of forms into a short list of physical changes, and lets the team check later whether each change worked.
Does near-miss monitoring mean watching individual workers?
It should not. A well-designed system reports time by crew, and its supplier can tell you exactly which screens show workers’ names and which look-ups of one person need a stated reason and are written to a log. Crews report more when they trust the system, so clear answers on names and look-ups are a safety feature as well as a fairness one, and they should be checked before purchase.
