Port & Terminal Operations

The problem: ambient intelligence at ecosystem scale

A container terminal has to coordinate ship-to-shore cranes, yard cranes and straddle carriers, a fleet of internal and external trucks, rail sidings, berth scheduling, and the weather and tidal conditions that determine when a vessel can safely dock. Every one of those systems typically arrives from a different vendor, on a different protocol, updated on a different cadence — crane telemetry in near real time, weather feeds every few minutes, vehicle position continuously, gate and yard-management events on their own schedule.

The usual approach is a separate integration for each system, each with its own gateway, its own licence, and its own point of failure. That works until an operator needs to correlate across them — for example, holding a crane movement because a truck is still in the bay, or adjusting berth scheduling because a tidal window closed early. Correlation across systems that were never designed to talk to each other is where most terminal automation projects get stuck.

How Relay fits

• →One platform, every protocol: crane and yard-equipment telemetry over Modbus or OPC UA, vehicle position over MQTT or cellular, and environmental feeds over whatever the sensor vendor shipped — all land on the same Relay cluster with no per-protocol licence and no gateway sidecar per vendor.
• →Real-time routing and correlation: Relay's built-in filtering and rules let an operator express "hold this crane movement while that truck is in the bay" as a routing rule, not a bespoke integration project.
• →Massive connection scale: a terminal's combined fleet of cranes, vehicles, gate readers and sensors is well within Relay's 10M+ concurrent connection capacity, on one control plane.
• →Zero downtime: cluster mode with real failover means a terminal that never stops moving cargo doesn't have to schedule a maintenance window to upgrade the system coordinating it.

A typical architecture

Crane and yard-equipment controllers publish operational telemetry over their native industrial protocol. Vehicle-tracking systems publish truck and internal-transport position over MQTT or cellular. Weather and tidal feeds arrive over whatever the provider offers, ingested through Relay's HTTP and webhook support. A terminal-operating-system integration or automation agent subscribes to the combined stream, and Relay's routing rules apply cross-system logic — crane holds, berth adjustments, gate sequencing — without each source system needing to know about the others.

Why this scales past one terminal

Because Relay's rate card and protocol support don't change between a cloud region, an on-prem rack, or an edge site at the terminal itself, an operator running multiple terminals can standardize on the same platform at each site, with each terminal's cluster keeping local traffic local and only forwarding what a group-level system actually needs — rather than routing every crane movement through a single distant point of integration.