Project

PRJ-0196

Industry

Machine Building & General Manufacturing

Application

SCADA & production data

Commissioned

Nov 10, 2023

Status

Supported

OEE measured on 14 machines and raised from 58.1 % to 67.0 % once downtime reasons were coded at the machine.

Project

PRJ-0196

Industry

Machine Building & General Manufacturing

Application

SCADA & production data

Commissioned

Nov 10, 2023

Status

Supported

OEE measured on 14 machines and raised from 58.1 % to 67.0 % once downtime reasons were coded at the machine.

Production data architecture across the OT/IT boundary — FIG. 1
SCADA & production data
PRJ-0196

Commissioning results

Measured outcome

Commissioning results

Measured outcome

OEE

58.1

→

67.0

14 machines · months 4–6 vs. first 4 weeks of data

OEE

58.1

→

67.0

14 machines · months 4–6 vs. first 4 weeks of data

Availability

→

Same period · performance 84 → 88 %, quality 96 → 96.4 %

Availability

→

Same period · performance 84 → 88 %, quality 96 → 96.4 %

Stops with a reason code

Stops longer than 2 min · months 4–6

Stops with a reason code

Stops longer than 2 min · months 4–6

Weekly reporting effort

→

0.5

h/week

Production engineering · spreadsheets replaced by dashboards

Weekly reporting effort

→

0.5

h/week

Production engineering · spreadsheets replaced by dashboards

Engineering record

Challenge, approach and architecture

Engineering record

Challenge, approach and architecture

Challenge

Context. A plant making machined and assembled components on 14 machines — CNC centres, presses and assembly stations — built between 1996 and 2021.

Existing process. OEE was estimated weekly from operator logs and ERP bookings. Engineers spent about six hours a week building reports, and the largest losses were argued about rather than measured.

Constraints.

Four controls generations, two PLC platforms and three machines without a usable controller interface.
A flat plant network shared with office IT.
No changes to machine safety or machine programs.

Challenge

Context. A plant making machined and assembled components on 14 machines — CNC centres, presses and assembly stations — built between 1996 and 2021.

Constraints.

Four controls generations, two PLC platforms and three machines without a usable controller interface.
A flat plant network shared with office IT.
No changes to machine safety or machine programs.

Engineering approach

We segmented the network first, then connected machines through an edge gateway per area, and agreed a reason-code tree with operators before building any dashboard.

Scope. Network audit and segmentation with an IT/OT firewall; OPC UA connectivity for PLC-controlled machines; I/O modules for the three machines without an interface; Ignition SCADA with historian, OEE calculation and a downtime-coding screen at each machine; weekly reports generated automatically.

Engineering approach

We segmented the network first, then connected machines through an edge gateway per area, and agreed a reason-code tree with operators before building any dashboard.

System architecture

Network: separate OT VLAN with firewall to plant IT, documented addressing.
Edge: 3 gateways with local buffering, OPC UA to machines, MQTT to the server.
Server: Ignition with SQL historian, OEE model and reports.
Machine screens: 14 tablets for reason coding, maximum 2 taps per stop.

System architecture

Network: separate OT VLAN with firewall to plant IT, documented addressing.
Edge: 3 gateways with local buffering, OPC UA to machines, MQTT to the server.
Server: Ignition with SQL historian, OEE model and reports.
Machine screens: 14 tablets for reason coding, maximum 2 taps per stop.

Deployment & commissioning

Machines were connected in three groups over six weeks without production stops. The first four weeks of data served as the measured baseline; reason coding was introduced from week five, and improvement actions from month two.

Deployment & commissioning

PLCMachines14, mixed generations
EDGEGatewaysOPC UA, buffered
MQTTTransportThrough OT firewall
HISTHistorianSQL + OEE model
DASHDashboardsShift and weekly

FIG. 3 · Signal path · Rev A

Specification & results

Specification, acceptance and results

Specification & results

Specification, acceptance and results

Machines

4 controls generations

Machines

4 controls generations

Gateways

Local buffering

Gateways

Local buffering

Protocols

OPC UA · MQTT

Protocols

OPC UA · MQTT

Reason codes

3-level tree

Reason codes

3-level tree

Data latency

Machine to dashboard

< 2 s

Data latency

Machine to dashboard

< 2 s

Parameter

Specified

Measured

Result

Machines connected

Pass

Machines connected

Specified

Measured

Result

Pass

Data latency

< 5 s

< 2 s

Pass

Data latency

Specified

< 5 s

Measured

< 2 s

Result

Pass

Gateway buffering

≥ 24 h

72 h

Pass

Gateway buffering

Specified

≥ 24 h

Measured

72 h

Result

Pass

Stops coded

≥ 90 %

92 %

Pass

Stops coded

Specified

≥ 90 %

Measured

92 %

Result

Pass

Results

Measure	Baseline	Months 4–6
Availability	72 %	79 %
Performance	84 %	88 %
Quality	96 %	96.4 %
OEE	58.1 %	67.0 %

OEE = availability × performance × quality: 0.72 × 0.84 × 0.96 = 58.1 %; 0.79 × 0.88 × 0.964 = 67.0 %. Most of the gain came from two causes the data made visible: waiting for material on four presses and unplanned tool changes on two CNC centres.

Results

Measure	Baseline	Months 4–6
Availability	72 %	79 %
Performance	84 %	88 %
Quality	96 %	96.4 %
OEE	58.1 %	67.0 %

Drawings

Layout and detail drawings

Original technical drawings prepared for this record. Illustrative, not to scale.

Drawings

Layout and detail drawings

Original technical drawings prepared for this record. Illustrative, not to scale.

FIG. 2 · OEE breakdown, baseline vs. after

We stopped debating whose numbers were right. The meeting is now about the top three losses.

Plant Manager

Client · industrial components (anonymized)

PRJ-0196

Solutions, platforms and resources

Industrial Networking & Production Data

Machine connectivity, OEE and downtime data across mixed controls generations, with a defined IT/OT boundary.

→

SOL-08

Industrial Networking & Production Data

Machine connectivity, OEE and downtime data across mixed controls generations, with a defined IT/OT boundary.

→

SOL-06

HMI & SCADA

Operator screens and supervisory systems designed around the decisions operators actually make.

→

SOL-06

HMI & SCADA

Operator screens and supervisory systems designed around the decisions operators actually make.

→

Technology stack

HMI / SCADA

TEC-12

Ignition by Inductive Automation

SCADA, HMI and production-data applications built on Ignition, including OEE and downtime reporting.

OPC UA · MQTT · SQL

Experience

IIoT / Edge

TEC-13

OPC UA & MQTT

Vendor-neutral machine connectivity through edge gateways, with buffering and a defined IT/OT boundary.

OPC UA · MQTT · Sparkplug B

Integrates With

PLC / PAC

TEC-01

Siemens SIMATIC S7-1500 / TIA Portal

PLC and safety-PLC programming, TIA Portal migrations from legacy S7 and S5 platforms, PROFINET network design.

PROFINET · PROFIsafe · OPC UA

Experience

PLC / PAC

TEC-02

Rockwell Automation Allen-Bradley ControlLogix / CompactLogix

ControlLogix and CompactLogix control and safety programs, PLC-5 and SLC 500 migrations, FactoryTalk integration.

EtherNet/IP · CIP Safety

Experience

Related resource

Application note

Why OEE data needs reason codes before dashboards

Production data

Article · 5 min read

Read

→

Why OEE data needs reason codes before dashboards

Production data

Read

→

Have a similar application?

Discuss a production-data project

Start a Project Brief

→

All project records

→

An engineer replies within one business day.

Have a similar application?

Discuss a production-data project

Start a Project Brief

→

All project records

→

An engineer replies within one business day.