An industrial operation has many moving parts. However, the traditional way of monitoring industrial systems is limited. If you can’t see what’s happening "under the hood" of your production - it’s very difficult to improve!

Costly production problems in an industrial operation are often hidden. The problems first have to be identified, and their causes need to be understood. Without a proper observation model, issues with your production lie latent, unreported, and undiagnosed. These issues emerge as costly problems when you least want them to. To solve this. We’re proposing is a dynamic “map” of your value chain. A whole-of-system model of your industrial operation, which is dynamised by real-time information streaming from your sensors.

Our solution has four steps: observing, analysing, visualising, and enabling. It is designed to be scalable and simplify the ease of connectivity between applications and devices.


Collecting data is more of a cultural hurdle than a technical one. In most industrial use cases, sensors are quite cost-effective relative to the value of the assets they’re monitoring.

With the Unified Namespace (UNS) approach, your industrial operation can be mapped out cohesively. The UNS is a dynamic repository which ingests and contextualises the data streaming out of your assets across your business. Any application or device can publish or subscribe data to the UNS.

The UNS approach facilitates the interconnection of data across historically distinct systems, including:

  • Instrumentation Data from sensors, SCADA, or machines.
  • Visual data from cameras.
  • Positioning data from mobile assets.
  • Text or tabular data across your enterprise systems (e.g. databases, ERP, WMS, or CRM).
  • Machine Learning Algorithms.

With the UNS regime, information from any node (e.g. fixed plant, mobile plant, enterprise systems, cameras) is accessible to any other node. Each node becomes both a producer and a consumer of information. Information from any asset is accessible from any other asset, in a standardised way.

This UNS approach enhances the scalability and ease of connectivity between your industrial devices and application. Additionally, by “decoupling system dependencies” the UNS significantly reduces the cost and time to develop and integrate new capabilities.

The unified namespace approach enables you to observe your plant by gathering and compiling real-time data across your organisation into an accessible format. Ultimately, improving situational awareness by using data.


Once your data is streaming into the unified namespace, the process of analysing it is simplified.

Each source of raw data is a variable which describes your operation. However, no individual sensor reading tells you the full story.

Combinations of these individual variables and events (e.g. sensor readings, camera detections, production events) can be fused to form numeric descriptors of your operational state. The fusion of multiple data sources yields more consistent, accurate, and useful information than that provided by any individual data source.

This analysis enables you to uncover hidden patterns in your data, map it to meaningful indicators on your production, and make corresponding data-driven changes to improve your operations.

Our speciality is the development of learning algorithms which ingest all of these variables, and output meaningful indicators of your production.
Learn about our machine learning services here.


In much the same way that a map has different layers, like terrain, streets, satellite imagery, a visualisation increases situational awareness by displaying layers of your industrial operation visually.

A digital twin is a visual depiction of your operation, digitally illustrating your operations so that you have greater clarity of the current operating scenario.

Visual patterns can be quite effective at conveying the operational state:

  • Colours can be used to show the current state.
  • Paths can be used to display the trajectory of your mobile plant as it progresses through your production floor.
  • Markers can be used to indicate the presence, type, and location of an alert item.


With each node in a UNS-enabled operation being both a producer and a consumer of information, it becomes simpler and more cost-effective to add smarter production capability.

Data and insights can Traverse the network, with devices and applications publishing or subscribing to topics that are relevant to them. This information architecture enables your production to become more adaptive, as your systems can dynamically alter their configuration based upon events, recognised patterns, and long-term trends. Ultimately, through increased resource flexibility, enables more profitable decisions to be made.

As flexible production capability is added, these devices tie into the same UNS, and gain the network effects of having operation-wide production information readily accessible to guide decision making.

Once the data interfaces are interconnected, your operation behaves more like an autonomous system. Optimisation algorithms can provide supervisory guidance by being tied to a predefined objective, such as minimisation of wait time in a multi-step fabrication sequence, or maximisation of human and robot resource utilisation in a cooperative process.

Final Thoughts

You can map out your operations through a process of obervation, analysis, visualisation, and enablement. This “map” of your value chain, is dynamised by your streaming real-time data, and enables you to see your operations at an expanse beyond your own physical field-of-view.

Data enabled manufacturers are going to outcompete their counterparts in terms variable cost per unit, as they can leverage their history of workflow data to enable smarter production.

There a sizable opportunity cost to not having operational data. To get more production output for the same amount of capital input, industrial businesses should be mapping out their operations.

Book a data discovery. We'll help you determine how your organisation can extract wisdom from its operational data.