Integrations

At Intellect, we believe in connecting workers. This means giving them access to the knowledge, data, and best practice that's stored across the organisation. Part of how we do this is by integrating Intellect with other systems.

This page covers the different types of integration available, the use cases for them, and technical detail on how they're built.

The first section is for anyone who wants to understand the use cases for integrations and the systems involved. The last two sections are a technical and security overview of how Intellect integrates.

1. Use Cases

This section outlines common systems we integrate with, and the use cases for those integrations.

1a. Master Data Management — ERP, MES

Data within Intellect is structured using the organisation's core master data. Because of this, we recommend synchronising master data with your core system of records via unique identifiers.

¹ E.g. work instructions, tags, CILs, breakdown reports, etc.
² E.g. batch records, quality incidents, hygiene audits, etc.
³ E.g. near miss, incidents, audits.
⁴ E.g. daily management systems, gemba walks and boards, kaizens.

1b. Alerts and Notifications — MES, IoT, SCADA

Within Intellect, it's possible to push alerts and tasks to users based on real-time events. Each event can be pushed into Intellect along with:

• • Who should be notified (either specific users or groups of users, such as "Line 1 Operators")
  • What users should do with the notification — view it as an FYI, confirm they've seen it and taken action, or complete a specific task related to it (e.g. filling out a breakdown report)
  • By when action should be taken
  • Meta information that goes with the alert, which can be shown directly to the worker

This can be used for a range of use cases. For example, an MES can trigger an alert when a machine goes down, prompting the operator to log a breakdown report in Intellect.

1c. Mobile Maintenance — ERP/PM

Using integrations, it's possible to extend existing plant maintenance suites to the frontline. In Intellect, users can create, receive, edit, and update work orders. Master data integrations also let users view and consume spare parts as part of completing a work order.

This brings legacy maintenance solutions to the frontline in a user-friendly way. Maintenance activity is displayed in one interface alongside other shop floor activities in Intellect, giving frontline employees a single to-do list and one pane of glass.

1d. Reporting and Analytics — BI Suite, Office Software

Data captured in Intellect can be pulled into other systems. Commonly this is pulled into reporting solutions such as Power BI, Tableau, or Yellowfin, or directly into Microsoft Excel for specific analysis.

Pulling operational data from Intellect into an analytics solution makes it possible to cross-reference frontline feedback with other data sources, such as MES or IoT solutions, for a complete picture.

Intellect supports all major analytics solutions, with the deepest support for Microsoft Power BI. With full support for Power BI embedded, dashboards can be embedded directly into Intellect, making them more accessible to workers.

1e. SOPs and Work Instructions — LMS, DMS, QMS

Within Intellect, it's possible to embed work instructions, SOPs, and training materials directly into a user's workflow. This content can be uploaded or created directly in the Intellect portal, but for many customers, it already exists in a DMS (e.g. SharePoint) or a QMS (e.g. Intelex). Intellect can integrate with these systems so the documentation shown to users stays automatically in sync with the source of truth.

By linking Intellect to these systems, workers can be confident they're following the correct procedures. As users complete tasks in Intellect, they're one click away from the approved work instructions or SOPs — helping upskill workers and reduce procedure non-compliance.

2. Integration Connectors

Intellect has developed a variety of connectors for these integrations, built on an integration framework that makes it easy to add new connectors as needed.

Our integrations are built around modern, widely adopted protocols — REST APIs, MQTT, and cloud services (AWS, Google Cloud, Azure, etc.). We generally recommend against direct PLC connections (e.g. Siemens, Omron, Allen-Bradley), legacy (non-UA) OPC, serial or raw TCP/UDP connections, FTP or Samba file shares, and email-inbox-based integrations, as these depend heavily on a customer's internal technical resources and typically add significant delivery time and risk. Deeper, machine- or SCADA-level interfacing sits outside standard Intellect integrations and is scoped separately, closer to a full MES implementation.

2a. REST APIs

Intellect has a suite of standard APIs for common tasks, used to build integrations between systems — this is our preferred method of integration. The APIs use a REST structure and return data as JSON over HTTPS. These are used when Intellect is the passive actor in an integration (though we're always happy to offer support).

2b. Custom API Endpoints

Sometimes we spin up new API endpoints for specific purposes — for example, if an existing integration posts data in a specific format, we can create a new endpoint that slots in using the same format.

2c. API Push

When it makes more sense for Intellect to be the active actor in an integration — such as pushing data submitted in Intellect to another system — Intellect can call external APIs. These calls can run on a fixed frequency or be triggered by events in Intellect, such as every time a user completes a task.

2d. Flat File Transfer

Although we don't normally recommend it, in some cases it makes sense to transfer data via flat files on a shared file server. We support SFTP for this, typically using CSV as the file format, though we can be flexible.

2e. SQL Database Access

Sometimes customers want to pull data from Intellect into systems that don't support API endpoints. For this, we can spin up a database (MS SQL, PostgreSQL, or other) for access. We never give direct access to production databases — instead, data is replicated from production into a cloned analytics database.

2f. Real-Time Connectors

For use cases needing continuous, low-latency data exchange, Intellect supports several real-time connection methods:

• • HTTP API Streams — a persistent HTTP connection where data is continuously streamed from server to client.
  • WebSockets — a full-duplex connection enabling real-time, two-way data exchange.
  • MQTT — our preferred protocol for real-time communication in manufacturing integrations, often replacing older streaming or polling methods.
  • OPC UA — serves a similar purpose to MQTT, but is recommended for internal network connections rather than transmission across the public internet or network tunnels.

Where a message broker sits behind an API endpoint, common examples we work with include RabbitMQ (complex routing and reliable task distribution), Amazon SQS (managed point-to-point queuing), and Amazon SNS (managed pub/sub broadcasting to multiple systems).

3. Security

Intellect supports varying degrees of security for integrations, depending on a customer's security goals and the sensitivity of the data being transferred.

All APIs are secured using token authentication, SSL encryption, and user-level access control as standard. On top of this, we can offer additional security if needed.

3a. VPC Peering

To avoid making API endpoints publicly accessible, we support VPC peering — creating a secure link from the Intellect environment into the customer's VPC. All communication is encrypted and not publicly accessible.

3b. Edge Node Implementation

When data is being pulled from on-premise systems or equipment, it may not be preferable to expose those endpoints to third parties. To manage this, Intellect supports installing an edge node inside the customer environment, with restricted access to only the specific data required.


If you have any questions on Intellect's Integration capability and options, please contact your dedicated Intellect account team.