Overview
An industrial platform delivered and running in production for an automotive-components factory. I connected shop-floor automation with business operations, bringing PLCs, Modbus devices, OPC UA gateways, SCADA, production tracking, quality, maintenance, energy and SAP into one operational view. The result is a traceable digital thread from raw material and machine cycle to finished part, warehouse movement and delivery, supported by a real-time dashboard for plant managers and production teams.
A digital operating layer for an automotive-components plant
This project modernized the information flow of a factory producing precision components for the automotive industry. I connected production, maintenance, quality, logistics and management without replacing the systems already operating on the shop floor. The production platform created one reliable operational picture of the factory, from the first machine signal to the final business transaction.
Connecting PLCs, machines and enterprise systems
The architecture used industrial gateways to ingest data from PLCs, sensors and legacy equipment through Modbus TCP/RTU and expose normalized machine information through OPC UA. A SCADA layer remained responsible for supervision and operator interaction, while an edge service buffered, validated and enriched events before forwarding them to the central platform. SAP was connected through controlled APIs, OData and asynchronous business events for orders, materials, production confirmations, inventory and maintenance data.
One dashboard for the complete production system
The production dashboard combines live line status, machine availability, cycle time, rejects, alarms, work orders, production targets, energy consumption and material flow. Different views serve operators, maintenance, quality, plant management and the central operations team. Instead of forcing every role into the same interface, the system presents the same underlying truth at the level of detail needed for each decision.
From raw material to finished part
Each production order was linked to the material batch, machine, programme, operator shift, process parameters, inspections and final packaging. The platform could correlate an abnormal machine condition with the affected production window and identify which parts, lots or deliveries required review. This created a digital thread across MES-like production records, quality events, warehouse movements and SAP master data.
Turning plant data into action
Rules and event-driven workflows created maintenance tickets when vibration, temperature, cycle time or alarm patterns moved outside their expected range. Quality deviations could stop or quarantine a lot, notify the right team and attach the relevant machine context automatically. Forecasting and anomaly-detection models estimated downtime risk, energy peaks and process drift, while human approval remained part of the operational loop.
A resilient edge-first system for a factory that cannot stop
The system separated real-time plant communication from business workloads. Edge nodes continued collecting and buffering data during network interruptions, messages were published through an event bus, and central services exposed APIs for dashboards, analytics and SAP integration. Historian and time-series storage supported high-frequency telemetry, while PostgreSQL stored operational entities, permissions, traceability and workflow state. Observability covered gateway health, data freshness, message lag and integration failures.