The field of IoT has led to an unprecedented accumulation of data from industrial environments.
Location of vehicles via GPS, assets via RFID, proximity via BLE, state of equipment via sensors, and movements, assignments, or incidents recorded by connected systems are all data points.
The real engineering question then arises:
Data vs. Context: "A signal is not context."
And in industrial transportation settings, this distinction is critical.
A Location Event is Only the First Step
Imagine this report:
Vehicle: VH-204
Location: Yard B
Time: 14:32
This information alone has value.
However, operations teams would rather see information that translates directly into actionable intelligence, such as:
Is VH-204 available?
What assignment is it tied to?
Is it loaded? If so, with what?
Is the loading process complete?
Is the destination ready for arrival?
Is the necessary equipment accessible and in good working order?
Is a driver assigned to it?
Is a maintenance issue impacting its availability?
The first data set describes where something is.
The second dataset represents a situation in operations.
This is where context-aware systems show their worth.
Linking Things Together
An industrial transportation network can best be visualized as a series of interconnected entities and their relationships:
Vehicle
Assignment
Freight
Destination
In the real world, the relationship would likely look more like:
Vehicle Freight
Driver Assignment
Equipment Facility
Maintenance
Every entity in the network is a source of information.
The trick then is to relate those signals together.
An AIoT architecture has the potential to aggregate these signals in such a way that applications are no longer seeing only isolated events but holistic circumstances.
The Implications in the Industrial Realm
The unique operational realities of industrial transportation include large-scale physical locations and specialized assets.
Operations often integrate a diverse range of elements:
Fleets
Heavy equipment
Railcars
Containers
Workers
Freight
Loading equipment
Storage areas
Maintenance resources
Facilities
Traditional systems are accustomed to managing one or more of these categories but often not all of them and independently.
This creates a visibility blind spot.
For instance, a fleet application can track a vehicle's location, an asset-management system can identify a trailer's status, and a workforce application can reveal an employee's availability.
Operational questions, however, will require input from all three systems.
Using AI to Intersect Signals
IoT excels at information acquisition.
AI can be used to add another layer of value by understanding trends, correlations, anomalies and the operational context of data.
Here's a conceptual architecture:
Physical Environment
Sensors / RFID / BLE / GPS
Data Collection
Entity & Event Context
AI / Analytics
Operational Insight
Human Decision
The Goal of this system: make information useful for decision-making, not just to generate another data dump.
Heavy-Haul Transportation: A Use Case
In a heavy-haul operation that specializes in moving specialized equipment, various conditions will affect operations:
Vehicle readiness
Trailer readiness
Freight availability
Loading equipment at the operation
Worker availability
Access to the facility
Current assignments
On-road traffic/weather conditions
Maintenance issues
Tracking the location alone will not suffice. A context-aware system that combines these various signals can represent the operational state far more effectively. This same methodology can be applied to rail yards and container terminals.
Rail and Terminal Applications
The sheer volume of both static and mobile assets at rail yards and terminals implies a real-time operational environment where a working system must take into account relationships between locomotive engines, railcars, freight, tracks, yards, personnel, and facility operations. Container terminals share similarities, consisting of cargo containers, truck drivers, cranes, laborers, loading docks, storage areas and scheduling operations. In all situations, tracking location is critical but not sufficient, the interconnectivity of assets is also of paramount importance.
Asking the Right Questions to Avoid the "Dashboard Problem"
One important principle in building industrial IoT systems:begin with your questions.
What am I trying to do?
Do you want to know where an asset is currently located?
Is it available to move?
Why is something delayed?
Which resources are interacting right now?
What triggered a change?
Which of my devices are underutilized?
What's blocking the next operation from happening?
The questions can inform the type of signals and systems that will be required. Without this critical consideration in the early stages of designing a solution, companies will often end up creating a data pipeline without a clear operational focus.
AIoT for the New Generation of Transportation Systems
The next phase of industrial transportation IoT is the establishment of much deeper ties between raw, physical signals and intelligent, operational decision-making. This does not intend to replace existing systems but to create an intelligence layer to integrate and optimize the entire network. Aperture Venture Studio's Industrial Transportation Group offers an example of this approach for both fleets and rail, port and pipeline, and heavy transportation operations.
The engineering concept is simple: visibility is derived from collecting data, context from interconnecting it, and decision-making capabilities can arise from turning context into intelligence. In industrial transportation operations, these connections could very well determine the future effectiveness of the industry. For more info visit: apertureventurestudio.com
Top comments (0)