COMPO — SCADA Digitalisation

Project record

Optimising industrial and logistics processes through digital transformation and SCADA integration

The project record and its abstract, as submitted to UNET.

Year: 2025 — term 2024-3
Type: Professional Application Project (internship)
University: UNET — Industrial Engineering
Author: Jonathan José Rincón Márquez
Academic tutor: Ing. Bethy Pinto
Industrial tutor: Ing. Maxime Van Landeghem
Company: Saint-Gobain Glass — Thourotte, France
Pages: 149

The report describes the development, implementation and validation of a digital system to optimise the industrial and logistics processes of the COMPO department at Saint-Gobain Glass, Chantereine plant. The aim was to turn manual data handling into an automated, integrated environment, guaranteeing traceability and real-time control of raw materials and cullet.

A SCADA system was designed and implemented on the Ignition platform, connected to a Siemens S7-1200 PLC, industrial sensors and a MySQL database. Development followed an agile FDD methodology, which supported the progressive build-up of modules: users, suppliers, hauliers, materials, inbound orders, shipments and inventory consumption.

Dashboards and key performance indicators were integrated for dynamic process monitoring, together with automatic reporting. The tool fully replaced the manual records and consolidated a robust, scalable system aligned with Industry 4.0, Lean Manufacturing and the World Class Manufacturing model.

Industry 4.0 SCADA Ignition PLC MySQL OPC-UA Automation Industrial logistics Continuous improvement

Chapter I — The problem

The whole workshop's logistics lived in spreadsheets

Raw materials and cullet were logged by hand, transcribed into Excel files saved locally, and left to each operator's own judgement. No cross-validation, no central backup, no real-time view of stock. The diagnosis took the first five weeks: semi-structured interviews, direct observation and document analysis.

Truck on the plant weighbridge — **Where it starts** The truck weighbridge — the exact point where, until then, someone wrote the weight down on paper.

Stock tracking spreadsheet with formulas pointing to a network file — **Before** The stock tracking file — values and network path masked.

Diagram of the manual process in the COMPO department — **Existing process** From weighing the truck to updating stock, with three separate files along the way.

Chapter IV — The architecture

From the weighbridge to the database, with no one in between

The truck's weight leaves the PLC, passes through the MOXA converter, travels over OPC-UA and Modbus, is written to MySQL and appears in Ignition. That path used to be a note on paper; today it is a transaction.

IgnitionSCADA platform

Siemens S7-1200Weighbridge PLC and sensors

MOXASerial-to-Ethernet converter

OPC-UA / ModbusField protocol

MySQLSGG_CHT_COMPO database

Architecture diagram: weighing, MOXA, PLC, OPC-UA Modbus, MySQL, Ignition — **Data chain** The six hops a weight reading takes from the weighbridge to the screen.

Ignition Designer, the project's development environment — **Build** The whole project was developed in Ignition Designer. Credentials masked.

MOXA converter configuration for the link to Ignition — **The link** MOXA converter configuration. IP address and serial number masked.

Entity-relationship model of the SGG_CHT_COMPO database — **Database** SGG_CHT_COMPO — users and security, logistics and transport, production, quality, reporting, suppliers, alerts and PLC data.

Materials table definition: columns, types and keys — **Documentation** Every table, column by column: type, nullability and key.

Chapter IV — The tool

Six profiles, one operational truth

Each role opens its own main window and sees only what belongs to it. Indicators change colour by value: stock below 15% turns red without anyone having to go looking for it.

Main window for the administrator role — **Admin** Users, suppliers, hauliers, materials and orders.

Main window for the manager role — **Manager** Team, reports, status, stock, production and shipping.

Main window for the operator role — **Operator** In, summary and out — three buttons, no ambiguity.

Main window for the logistics role — **Logistics** Materials, alerts and consumption reports.

Gatehouse window: truck entry and weighing — **Gatehouse** Truck entry, actual weight and the gap against what was declared. Operational data masked.

Consumption statistics window with a line chart — **Cumulative consumption** One of the dashboards: bar, line and pie charts, filtered by date, material, haulier or supplier.

Real-time silo level window — **Silo level** Material height is converted to tonnes by a Python script.

Material consumption window by period — **Consumption** By hour, by day and by week, material by material. Values masked.

Notifications window, filtered by role and department — **Notifications** Messages routed by role, department or person, straight from the tool.

Barges being loaded for cullet shipments — **Shipping** Cullet leaves for other plants in the group: every load generates its own QR-coded ticket.

Chapter V — Results and handover

What changed in the workshop

01
Manual records and scattered Excel files were eliminated.
02
Pulling a report went from 30 minutes to under 3 seconds.
03
Full data traceability, with fewer keying errors.
04
Consumption calculations, QR-coded tickets and stock validation, all automatic.
05
Real-time alarms for overweight loads, sensor failure or critical silo levels.
06
All logistics information centralised, with role-based access.

The last four weeks went to making the solution last: user and technical manuals in Spanish and French, hands-on staff training, and fortnightly feedback meetings. Both manuals were reviewed, approved and signed by the plant.

Cover of the systems and procedures manual — **Manual I** Systems and procedures.

Signed approval letter for the standards and procedures manual — **Approval I** Signed 31/07/2025. Signatures masked.

Cover of the information distribution manual for SMQF — **Manual II** Information distribution in SMQF.

Signed approval letter for the database manual — **Approval II** Signed 31/07/2025. Signatures masked.