AI Forklift Monitoring

Warehouse AI

Forklifts are among the most expensive moving assets in a warehouse and the source of its most serious safety risk, yet most operations have no real visibility into how they are actually used. Rubicon A-Eye watches forklift activity, location and proximity to people through the cameras you already have, turning utilisation and safety into something you can see and act on, not something you infer after an incident.

The business challenge

A warehouse fleet manager typically knows how many forklifts the operation owns, what they cost, and roughly how many hours they are scheduled for — but very little about what actually happens between clock-in and clock-out. Forklifts sit idle between tasks, take inefficient paths across the floor, and operate in close proximity to pedestrians in aisles and dock areas, all without any record beyond what a supervisor happens to observe in passing.

This absence of data creates two separate but related problems. The first is operational: without visibility into actual utilisation, fleet sizing and shift planning are based on assumption rather than evidence, and it is impossible to tell whether congestion on the floor is caused by too few forklifts, poor routing, or simply uneven task allocation. The second is safety: forklift and pedestrian interactions are one of the most common sources of serious incidents in warehouse environments, and near-miss events — situations that almost became an incident but did not — go almost entirely unrecorded unless someone happens to witness one and chooses to report it.

Day to day, this shows up as a recognisable set of problems for warehouse and safety managers:

• No reliable data on how much of a forklift’s shift is spent actively working versus idle
• Fleet size and shift staffing decisions made on intuition rather than measured demand
• Near-miss incidents between forklifts and pedestrians that go unreported and unrecorded
• Congestion and inefficient routing on the floor that nobody can diagnose without watching the whole shift
• Safety incidents investigated after the fact with only witness accounts to reconstruct what happened
• No way to identify which zones of the warehouse carry the highest risk of forklift-pedestrian interaction

The hidden cost of unmonitored forklift operations

An idle forklift is a sunk cost that keeps accruing — lease or depreciation, maintenance, and the labour cost of an operator who is being paid regardless of whether the truck is moving. Multiplied across a fleet of several units over every shift, low utilisation that nobody can see or correct represents a meaningful and entirely avoidable cost. On the safety side, the absence of near-miss data means that risk patterns — a particular intersection, a specific time of day when forklift and foot traffic overlap most — stay invisible until an actual incident forces the operation to look. By then, the cost is no longer hypothetical: it includes injury risk, potential downtime, investigation time, and the liability exposure that comes from not having addressed a pattern that, with visibility, would have been obvious in advance.

How Rubicon A-Eye solves the problem

A-Eye treats forklift activity as a continuous operational and safety signal rather than something assessed only through periodic observation. Cameras already covering the warehouse floor, aisles and dock areas feed an on-site edge device that identifies forklift location, movement and idle periods, and detects proximity between forklifts and pedestrians in real time.

All processing happens on a device inside your facility. Video never leaves the site; only the operational data it produces — utilisation states, movement patterns, proximity events — flows onward to dashboards and, where relevant, Odoo. This keeps the system aligned with data-privacy expectations, and because the focus is on equipment activity and safety conditions rather than identifying individual operators, it supports an operational improvement programme rather than personal surveillance.

In practice, this gives supervisors two things they did not have before. First, a measured view of how the fleet is actually used across a shift — which units are active, which are sitting idle, and where time is lost to congestion or inefficient routing. Second, a live safety signal: when a forklift and a pedestrian come into close proximity in a zone where that should not normally happen, an alert is raised immediately, supporting intervention before a near-miss becomes an incident rather than analysis of an incident after it has already occurred.

Key capabilities

What you can measure

• Active versus idle time for each monitored forklift across a shift
• Fleet-wide utilisation trends by shift, day and zone
• Near-miss proximity events between forklifts and pedestrians
• High-risk zones where forklift-pedestrian interaction occurs most often
• Routing and congestion patterns across the warehouse floor
• Idle-time incidents during periods of active task demand

Industry applications

Forklift monitoring matters anywhere powered material-handling equipment shares space with people and represents a meaningful share of operating cost. In the UAE and wider GCC, it is particularly relevant for:

For large distribution centres running multi-shift operations, utilisation visibility supports fleet sizing decisions that directly affect lease and maintenance spend. In any facility where forklifts share aisles with pickers and other foot traffic, near-miss detection targets the single highest-risk interaction on the floor. In operations with seasonal volume swings, utilisation data helps decide when extra units are genuinely needed versus when existing capacity is simply under-used.

Business benefits

• Evidence-based basis for fleet sizing and shift staffing decisions
• Earlier detection of near-miss situations before they escalate into incidents
• Reduced idle time through visibility into how the fleet is actually used
• Clearer picture of which floor zones carry the highest forklift-pedestrian risk
• A factual record to support safety programmes and incident investigations
• Better-informed routing and layout decisions based on observed traffic patterns
• Reduced reliance on after-the-fact witness accounts when investigating safety concerns

How it works: the operational workflow

The path from a camera view of the floor to a supervisor’s action is continuous, so utilisation and safety signals are available throughout the shift, not reconstructed afterward.

Example scenarios

Integration with your systems

Forklift utilisation and safety data delivers its value once it reaches fleet, safety and operations management directly. As a certified Odoo partner and AI engineering team, Rubicon connects this data into the systems you already run.

Frequently asked questions