Large industrial and utility enterprises are shifting from passive equipment monitoring to proactive management. In this model, industrial Internet of Things (IIoT) data becomes a direct trigger for business process automation, minimizing infrastructure downtime. However, in practice, Chief Operating Officers (COO) and Chief Technology Officers (CTO) often face a gap between operational technology (OT data from SCADA systems or sensors) and corporate ERP/BPM systems. This leads to slow response times to failures and excessive reliance on the human factor.
The effectiveness of IIoT lies not in accumulating terabytes of raw telemetry, but in the automatic conversion of critical signals into structured business processes—for example, maintenance work orders. This requires building a robust architecture and a secure gateway between OT and IT segments.
Why passive monitoring fails: the problem of isolated data in the industrial circuit
Raw sensor data often remains isolated within the industrial circuit. Direct integration of the technological network with corporate IT systems is blocked at the architectural level due to protocol incompatibility and security requirements. If an operator is forced to manually record deviations from a SCADA panel and enter them into an ERP, the main advantage of IIoT—speed of response—is lost.
Here are several typical scenarios illustrating the complexities of industrial sites:
- Isolation of legacy equipment: Enterprises often operate units whose control systems cannot be updated. According to the NIST SP 800-82 Guide to OT Security, such segments must be strictly isolated.
- Excessive traffic: A vibration sensor can generate thousands of readings per second. Sending all telemetry directly to the corporate cloud leads to network congestion.
- Process gaps: Receiving a signal about a critical engine temperature deviation should automatically create a maintenance work order, rather than simply lighting up an indicator on a dispatcher's panel.
Architectural pattern: distribution between edge and cloud
According to the principles of the AWS Well-Architected IoT Lens, the reliability of an IoT solution is established during the design phase of the "device → edge → cloud" architecture. The key task is the correct distribution of which data to process locally (at the edge of the network) and which to transmit to the cloud or a central data center.
At the level of edge controllers, continuous collection of physical parameters occurs. Simple threshold triggers are configured here to detect anomalies, allowing for the avoidance of reliance on complex and expensive AI models where clear mathematical logic suffices. If a parameter exceeds the norm, an event is formed and transmitted through a security gateway to the corporate circuit to generate a work order.
Security at the IT and OT intersection: NIST SP 800-82 and ISA/IEC 62443 standards
Cybersecurity for industrial systems is a continuous risk management process, not a final state. According to the NIST SP 800-82 standard, in OT environments, availability and continuity of equipment operation are higher priorities than confidentiality (unlike in classic IT). This requires specific approaches.
A basic control is the segmentation of IT and OT networks, which helps protect critical infrastructure. Communication between corporate and industrial systems must be built through a demilitarized zone (Industrial DMZ) in accordance with the requirements of the ISA/IEC 62443 framework. No device from the IT network should have direct access to controllers; data exchange is carried out exclusively through secure buffer servers.
Data normalization and edge filtering: traffic control
To combine disparate data from equipment, the OPC UA (OPC Unified Architecture) standard is used. This is a platform-independent architecture that allows for the normalization of data from various industrial controllers before their transmission to analytical or SCADA systems.
Local edge filtering allows for the extraction of only significant events from the stream. Industrial networks are evolving: according to the Ericsson Mobility Report, 5G will become the dominant mobile access technology by subscription count by the end of 2027, which will significantly expand the connectivity capabilities of wireless IIoT sensors. However, filtering "noise" directly at edge gateways remains a critical requirement for resource conservation and avoiding corporate database overloads.
Practical implementation: integration through platform solutions
Implementing such an end-to-end process requires specialized platforms and engineering expertise. Within the Intecracy Group—an alliance of independent companies linked by partner agreements and share exchanges—this task is solved comprehensively, with the involvement of Softengi (custom IoT/embedded development and integration).
To collect data from sensors, normalize protocols (MQTT, Modbus), and manage cyber-physical environments at the network edge, the AZIOT Platform is used. Filtered critical events are securely transmitted to the corporate level, where a solution built on the UnityBase platform handles routing and business process orchestration.
UnityBase is a full-stack JavaScript low-code platform. Using a unified Domain metadata model integrates work with data, API, and business logic. Thanks to built-in access control mechanisms (RBAC, RLS) and detailed audit trails, UnityBase ensures secure event import and automatic generation of work orders in BPM/ERP systems. For high-load projects or critical infrastructure, the developer recommends applying Enterprise or Defence editions of the platform, which support advanced security mechanisms.
Step-by-step algorithm for integrating IIoT telemetry into corporate business processes
- Step 1: Collection and local filtering. Configuring devices (edge) for primary noise filtering and data stream normalization according to the OPC UA standard.
- Step 2: Secure transit. Transferring critical events through a demilitarized zone (DMZ) with strict IT/OT network segmentation in accordance with NIST SP 800-82 controls.
- Step 3: Routing. Processing normalized events via an integration bus (based on UnityBase), checking threshold triggers, and determining the incident type.
- Step 4: Task generation. Automatically creating and assigning a maintenance work order in the corporate ERP/BPM system based on the received trigger.
- Step 5: Feedback. Recording the completion of work by the maintenance crew and automatically updating the equipment status in the enterprise monitoring dashboard.
FAQ
How can data be safely transferred from an isolated SCADA network to a corporate ERP without the threat of cyberattacks?
Secure transfer is ensured through the use of an Industrial DMZ and strict IT/OT network segmentation according to ISA/IEC 62443 and NIST SP 800-82 standards. Direct communication between circuits is prohibited; data exchange occurs exclusively through intermediate buffers or unidirectional gateways.
What data from vibration and temperature sensors should be processed at the edge, and what should be sent to the cloud?
High-frequency streams are processed at the edge level for immediate noise filtering and anomaly detection. Only aggregated metrics and critical events that act as triggers for work orders are sent to the cloud or corporate environment.
How can legacy industrial equipment be integrated into a modern automated work order system?
For legacy equipment that cannot be updated or connected directly, external edge controllers are used. They read physical parameters, normalize them (e.g., using OPC UA), and securely transmit telemetry to an isolated network segment without interfering with the unit's operation.