Our IT/OT engineering approach
First, together with the plant team, we identify which machines, connections and data flows affect the work. Then we validate the design in a pilot area and put the migration and rollback steps in writing.
Six principles we follow when working on a production network
In an OT environment, a faulty change can stop the line. That is why we work by the following six principles in every project.
Survey first, then change
No design decision is made until the inventory and flow information have been compiled from existing records and an on-site review.
Reversible steps
A configuration backup, maintenance window and written rollback procedure for every change that affects production.
Measurable acceptance
Success criteria are written down before work begins; handover is done with measurements and a signed record.
Documented handover
Diagrams, inventories, rules and procedures are handed over in a form the maintenance team can read.
Shared scope
The scope is written jointly with the production, automation and IT teams. It is agreed who responds first when a device fails, who approves changes and who keeps the records.
Clear limits
We state up front the work we do not do; where necessary, we coordinate with the right specialist.
We look at the layers and boundaries together
We use the Purdue model as a reference for discussing layers, and ISA-95 for describing the tasks and information exchange between production and corporate systems. Neither is a mandatory physical topology on site; we define the target architecture according to the plant’s needs.
Dashed line: data flow (OPC UA / MQTT, middleware) · Solid line: control traffic (e.g. PROFINET) · Yellow line: passage restricted by a firewall. The diagram is representative; product names are given as examples and do not imply a manufacturer partnership.
Five stages from discovery to operation
Discovery
Interviews, on-site review, read-only device data and passive traffic observation.
Design
Topology, data flow, security boundaries and acceptance criteria.
Pilot
Implementation, measurement and feedback in one cell or line.
Implementation
A phased migration spread across downtime windows, with a rollback plan.
Handover and operation
Up-to-date topology and connection documents (as-built), training, and monitoring and maintenance recommendations after implementation.
Who does what?
The division of tasks in a typical project; it is clarified together for each project.
| Topic | Eretek | Plant automation / machine manufacturer | Plant IT |
|---|---|---|---|
| Network, segmentation, OT DMZ | Design, implementation, documentation | Approval of production impact, on-site support | Corporate network and security standards |
| PLC data access | Data contract, integration, testing | Changes to the PLC project | Server and user management |
| PLC application software | Not done | Owner and responsible party | — |
| Machine safety / CE / SIL | Out of scope | Machine manufacturer / competent body | — |
Project scope and responsibilities
We do not offer the following as services; when needed, we coordinate with the relevant specialist.
- Writing or modifying PLC application programs
- Machine CE conformity and machine safety assessment
- Functional safety (SIL) calculation and verification
- IEC 62443, ISO/IEC or TISAX certification and independent audits
- Official reseller, service or warranty services on behalf of a manufacturer
- Guarantees of results, duration or success rates
Let's discuss the right architecture for your production network
Let’s plan a technical survey based on your current situation, installed equipment and goals.