Learn Automation with MINATA #56: SCADA Monitoring — Turning Machine States Into Operating Information

SCADA Monitoring: Turning Machine States Into Operating Information

SCADA turns the scattered states of many machines into information you can decide from. But a system that pulls every tag onto a dashboard with no stated purpose produces a screen that is beautiful and useless: many numbers, few decisions.

Start from the operating question

Decide first which goal the system serves: reducing downtime, tracking energy consumption, or quality traceability. Each goal decides which data has to be collected, how it is aggregated and who looks at it. Do not pull every tag just because it is there — choose the data that answers the question you asked.

Data quality has to be designed in from the start

A chart is only trustworthy when the data behind it is. From the beginning, design consistent timestamps, a notion of data quality — good, doubtful, missing — the handling of lost connections, and access rights. If the network drops while the dashboard keeps drawing a straight line as though nothing happened, it is deceiving whoever reads it.

Aggregate first, drill down after

Good SCADA gives the overview first and the detail second. An overview page shows Ready, Run and Fault states, downtime and the most frequent alarms across a group of machines. When something looks wrong, the user drills down into the specific machine to find the cause. Making an operator read hundreds of tags and aggregate them mentally is a failure of design.

A worked engineering situation

A SCADA overview page aggregates Ready, Run and Fault, downtime and top alarms across several machines. When one machine's downtime rises, the supervisor drills down into it, looks at the recent alarms and the shift on which they occurred, and hands maintenance a specific job. The data carries timestamps and quality flags, so "the machine stopped" is never confused with "the SCADA connection dropped".

The scope of the data, the architecture and the access rights all follow the operating goal and the policy of the plant.

Common mistakes

Pulling every tag onto a dashboard with no stated goal.
No quality flags, so a lost network still draws a normal-looking line.
Inconsistent timestamps, so nothing can be compared over time.
Detail pages only, with no overview to see the whole picture.
No access control, so anybody can change the monitoring configuration.

SCADA monitoring checklist

[ ] The monitoring goal was defined before the tags were chosen.
[ ] Timestamps, data quality and lost-connection handling were designed in.
[ ] There is an overview page and a drill-down into individual machines.
[ ] Rights to view and to change the configuration are clearly separated.
[ ] Charts stay honest when data is missing or doubtful.

Good monitoring turns data into decisions rather than putting numbers on display. #57 goes into panel energy metering — from one kilowatt-hour to a decision that reduces consumption.

Read more automation knowledge at MINATA: https://minatavn.com/en/blog/industrial-automation

Previous — #55: HMI alarm design: https://minatavn.com/en/blog/automation-55-hmi-alarm-design

Next — #57: Energy metering in the control panel: https://minatavn.com/en/blog/automation-57-panel-energy-metering

View all MINATA technical articles