Near-miss detection for warehouse and factory movement

Why near misses are hard to see in warehouses

Near misses are often missed because they happen inside changing mixed-traffic environments where people, forklifts, trucks, automated vehicles and temporary obstacles interact faster than manual observation can capture.

Near misses in warehouses and factories are often difficult to document because they happen inside busy, changing environments. Forklifts, pedestrians, trucks, manual handling equipment and automated vehicles may share aisles, crossings, loading areas and temporary storage zones. A close call can appear as a single moment, but the underlying pattern may be repeated many times.

Manual reporting, safety walks and incident reviews remain important. The challenge is that they usually capture what someone noticed, remembered or had time to document. In a high-traffic warehouse, many interaction patterns are easy to miss: a pedestrian route that repeatedly crosses a forklift path, a congested intersection during a specific shift, an obstacle that narrows an aisle, or a loading zone where speed variation changes throughout the day.

Near-miss detection for warehouse movement adds a measured layer of evidence. It helps teams review where risk indicators appear, how often certain movement patterns repeat, and which zones should be evaluated in more detail. The purpose is not to declare every event as a certified incident. The purpose is to support better decisions with objective movement insight.

Detect risk indicators with 3D movement analytics

3D movement analytics can help teams review positions, paths, direction changes, speed variation, traffic intensity, repeated interactions, congestion and time-to-collision indicators when the project setup supports that analysis.

Viscando’s approach is based on 3D and AI movement analytics. The system can help teams understand how different moving assets behave across a site, including people, forklifts, trucks, AGVs and AMRs. That makes it possible to review patterns that are difficult to see from a floor plan or a single observation.

Relevant signals can include:

• Positions and movement paths through aisles, intersections and shared areas.
• Direction changes and route overlaps.
• Speed variation and traffic intensity during different time windows.
• Repeated forklift, pedestrian, truck, AGV and AMR interactions.
• Congestion, waiting and dwell patterns.
• Obstacles, blocked aisles, buffer zones and emergency-exit visibility.
• Time-to-collision indicators where the project setup supports that analysis.

These signals can help EHS, HSE, warehouse, logistics and automation stakeholders move from general concern to specific questions. Which intersections create repeated attention points? Where do forklifts and people most often share space? Which areas should be reviewed before adding automation, changing routes or updating routines? Which changes should be measured again after implementation?

Analytics should be treated as decision support. It can reveal risk indicators and close-call patterns, but it does not replace risk assessments, safety training, supervision, procedures or human judgement.

From close-call patterns to safer decisions

The value is not one isolated data point; it is the ability to see whether interaction patterns repeat over time and whether they change after a layout, routing, training, process or automation decision.

The strongest use of near-miss analytics is not one isolated data point. It is the ability to see whether certain interaction patterns repeat over time and whether those patterns change after the site takes action.

1. Measure current movement in the relevant warehouse or factory area.
2. Identify recurring close-call indicators, conflict zones and mixed-traffic patterns.
3. Review the findings with EHS, operations, logistics and automation stakeholders.
4. Prioritize layout, marking, routing, training, process or automation changes.
5. Measure the same zones again and compare the before-and-after patterns.

This supports Viscando’s Measure → Improve → Evaluate approach. Teams can use the data to create a shared view of what is happening on the floor, decide which areas need attention first, and evaluate whether the next measurement period shows a different movement pattern.

Forklift near-miss detection and interaction analysis

Forklift near-miss detection is a strong use case because forklifts often share space with people, trucks, manual handling equipment and automated systems around intersections, aisle entrances, doors and loading areas.

Forklift near-miss detection is one of the clearest use cases for movement analytics in internal logistics. Forklifts often operate close to people, trucks, manual handling equipment and automated systems. Risk indicators may concentrate around intersections, aisle entrances, doors, loading areas, blind spots, temporary storage areas and shared pedestrian routes.

Viscando can help teams review where forklift interactions occur repeatedly and which movement patterns should be discussed in a safety or operations review. Useful questions include:

• Where do forklifts and pedestrians most often share space?
• Which crossings or aisle segments show repeated interaction patterns?
• Are there time windows with higher traffic intensity, congestion or waiting?
• Do route overlaps change when material flows, shifts or staging areas change?
• How do AGVs, AMRs and manual traffic interact with forklift and pedestrian movement?
• Which areas should be measured again after a layout or process change?

Use near-miss analytics for before-and-after evaluation

Short answer: by measuring a baseline and then measuring the same zones after a change, teams can compare movement patterns and decide what should be reviewed next.

Near-miss analytics becomes especially useful when a team wants to evaluate a change. Warehouses and factories often adjust pedestrian routes, forklift paths, aisle markings, staging zones, buffer areas, shift routines or automation plans. Without a baseline, it is difficult to compare what changed.

By measuring movement patterns before and after an intervention, teams can review whether the observed data supports the next decision. For example, a site may want to compare interaction density at an intersection, assess whether a route change moved congestion elsewhere, or evaluate how AGV and AMR movement affects existing forklift and pedestrian flows.

This before-and-after view can support internal alignment. EHS teams can discuss risk indicators. Operations teams can assess flow and congestion. Automation teams can review mixed-traffic behaviour. Management can see why an improvement area was prioritized and what should be reviewed next.

The data should be used carefully. It supports a more structured evaluation, but it does not prove a guaranteed safety outcome without approved project evidence and professional review.

Privacy-aware safety analytics

Short answer: near-miss analytics should be framed as movement-pattern evidence and decision support, with project-specific review for data handling, retention, employee trust and legal wording.

Near-miss detection often raises practical questions about cameras, employee trust and data handling. Those questions should be answered clearly and carefully.

Viscando’s value is movement analytics: turning observations of people, vehicles and space usage into data, visualizations and decision support. The focus is on understanding movement patterns and risk indicators, not on turning the warehouse page into surveillance messaging.

For implementation, the practical questions are project-specific: which areas are measured, what signals are relevant, who needs access to dashboards or reports, how findings are communicated internally, and how the data supports the safety improvement process.

FAQ