High performance HMI: screens that show the operator what is abnormal, the display hierarchy, and the ISA-101 life cycle
What makes a control system screen show abnormal conditions early: grey backgrounds and colour saved for alarms, analogue indicators against the normal range, the four level display hierarchy, alarms on the screen, the HMI life cycle from philosophy and style guide to screens under ANSI/ISA-101.01 and IEC 63303, and how to move an existing system across. With a drawing of the hierarchy and an analogue indicator.
Key takeaways
- A high performance HMI is designed to show the operator what is abnormal, not to draw the plant. Grey is normal, and colour appears only when something needs attention.
- Values are shown against their normal range and alarm limits with analogue indicators, so the operator sees at a glance whether a number is good rather than having to remember it.
- Screens are arranged in a hierarchy of four levels, from one overview of the operator's whole area down to the detail of a single loop, and laid out around the operator's tasks rather than the P&ID.
- ANSI/ISA-101.01 sets out the HMI life cycle, from a site philosophy and style guide through a reusable object library to the screens. IEC 63303 is the international standard derived from it.
- Most of the benefit comes from the philosophy, the style guide and the overview screen. Converting every screen in a single project is rarely needed to get it.
A reactor temperature climbs slowly over a night shift. The value is on the screen the whole time, a white number on a black background beside a dozen others, and nothing changes colour until it reaches the high alarm. The operator responds to the alarm, as trained. Nobody noticed the hour of drift before it, because nothing on the screen said the number was moving away from where it should be.
This guide covers high performance HMI design: what makes a screen show abnormal conditions early, the principles set against a conventional screen, the display hierarchy, the HMI life cycle that ANSI/ISA-101.01 and IEC 63303 describe, and how to move an existing control system across without rebuilding every screen at once. It is the screen side of alarm management, which covers the alarms themselves. The standards named are reference points, and each site's philosophy sets its own rules.
Showing the operator what is abnormal
A conventional control screen draws the plant. Vessels, pumps and pipes are coloured and often rendered in three dimensions, values are numbers beside them, and colour is used for equipment status. The plant is easy to recognise, and the important information is hard to find. The operator has to remember the normal value of every number and notice when one changes.
A high performance screen draws the situation. The plant is shown in simple grey outlines, values are shown against their normal range, small trends show which way things are moving, and colour appears only when something is abnormal. The operator does not have to remember whether 73 degrees is good, because the indicator shows it sitting in its normal band. The aim is to see a problem as it develops, before it reaches an alarm.
On a well designed screen, a normal plant looks boring. Colour is saved for the moment something needs the operator.
The principles side by side
The difference is a set of consistent choices rather than a single feature. Each one takes something away from the screen so that what is left carries more meaning. Status colour is the clearest case. Some industries use red for running and green for stopped, others the reverse, and sites with both conventions are common, which is one reason high performance designs keep status in grey and save colour for abnormal conditions.
| Element | Conventional screen | High performance screen |
|---|---|---|
| Background | Black, or a coloured or three dimensional rendering of the plant | A muted grey, so that colour stands out when it appears |
| Colour | Used for equipment status, often red for running and green for stopped, or the reverse | Reserved for abnormal conditions and alarms, so colour always means something needs attention |
| Values | Numbers on their own, which the operator has to remember the normal range for | Analogue indicators that show the value against its normal range and alarm limits |
| Trends | On a separate trend screen, opened when something is already wrong | Small trends built into the overview, so a drift is seen before it alarms |
| Equipment pictures | Detailed three dimensional vessels, pumps and animated flows | Simple two dimensional outlines that do not compete with the data |
| Layout | Copies of the P&IDs, laid out the way the engineer drew the plant | Laid out around the operator's tasks, with related values grouped |
| Navigation | A flat menu of screens, often dozens deep | A hierarchy from the overview to the detail, reachable in a few clicks |
| Alarms | Flashing colour wherever an alarm is configured | Priority shown the same way everywhere, by colour and by shape, so it reads even for a colour blind operator |
The display hierarchy
Screens are arranged in levels, from an overview of the whole area down to the detail of one loop. The overview is the most important screen in the system and the one most often missing. It is designed to be watched, not used, and shows the few values that tell the operator whether the area is running well, as trends and analogue indicators, with the alarm count by priority.
| Level | What it shows | When it is used |
|---|---|---|
| L1 Area overview | The operator's whole span of control on one screen: key values as trends and analogue indicators, and the alarm count by priority | Watched all shift to see whether anything is drifting |
| L2 Unit control | One unit or process, with the controls and values needed to run it | Where most routine operating happens |
| L3 Equipment detail | One piece of equipment or control loop in detail, with its full set of values and states | Diagnosing a problem and less frequent tasks |
| L4 Support | Interlock logic, diagnostics, procedures and help | Troubleshooting, often with an engineer or technician |
Alarms on the screen
The HMI is where the alarm system meets the operator, and the two have to be designed together. Alarm priority should be shown the same way on every screen, by colour and by a shape or symbol, so that it reads for operators with colour vision deficiency and in poor light. Alarm colours are not used for anything else. The alarm philosophy and the HMI philosophy share these rules, which is why many sites write them together. Alarm management under ISA-18.2 and IEC 62682 covers how alarms are chosen and prioritised, and the HMI decides how they are seen.
The HMI life cycle
ANSI/ISA-101.01 treats the HMI as a system with a life cycle, not a set of screens drawn at the end of a project. It starts with a philosophy and a style guide written once per site, then an object library built in the control system's tools, and only then the screens. IEC 63303:2024 is the international standard derived from it. Like the control system design documents, the early documents are short and decide almost everything that follows.
| Stage | What it produces | When |
|---|---|---|
| HMI philosophy | The principles: what the HMI is for, the display hierarchy, how colour, alarms and navigation are used | Once per site, before any screen is drawn |
| Style guide | The rules that apply the philosophy: palette, fonts, symbols, indicator types and screen templates | Once per site, kept current |
| Object library | The reusable faceplates, symbols and indicators built to the style guide in the control system's tools | Once per platform, version controlled |
| Display design | The screens themselves, built from the library around the operator's tasks | Each project, reviewed with operators |
| Testing and training | Screens tested against the task analysis and operators trained on them | Before the screens go live |
| Operation and audit | Feedback from operators, changes under change control, and audits against the style guide | For the life of the system |
Moving an existing system across
Few sites can rebuild every screen in one project, and few need to. The philosophy and style guide come first, then the overview screens, which give the largest share of the benefit for the least work. Unit and detail screens are converted as they are next touched, through upgrades, modifications and control system migrations. Operators should be involved from the first draft and trained before anything goes live, because a screen they do not trust will be worked around.
- Write the HMI philosophy and style guide, ideally alongside the alarm philosophy.
- Build the object library once, on the control platform, under version control.
- Design the level 1 overviews with the operators, around what they watch for.
- Convert level 2 and 3 screens as each area is next modified or upgraded.
- Audit new screens against the style guide before they are released.
- Manage every screen change through the site's change process.
Where HMI projects go wrong
- Screens converted to grey with no change to layout or indicators, so they are duller but no clearer.
- No level 1 overview, so the operator still has to page through screens to see the whole area.
- Alarm colours reused for equipment status or decoration.
- Each project or contractor building its own symbols instead of using the site library.
- Operators shown the new screens for the first time at cutover.
- The style guide written and then not enforced, so screens drift back over a few years.
Common questions
What is a high performance HMI?
A high performance HMI is a way of designing control system screens so that operators see abnormal conditions early. It uses a muted grey background, reserves colour for abnormal conditions and alarms, shows values with analogue indicators against their normal range, builds trends into the overview, and organises screens in a hierarchy around the operator's tasks. The approach was popularised by The High Performance HMI Handbook and is reflected in ANSI/ISA-101.01 and IEC 63303.
What is ISA-101?
ANSI/ISA-101.01-2015, Human Machine Interfaces for Process Automation Systems, is the ISA standard for the design, implementation and maintenance of HMIs. It describes an HMI life cycle starting with an HMI philosophy and a style guide, and covers display hierarchy, navigation, colour, alarm presentation and the management of the HMI over its life. It does not prescribe a particular look. IEC 63303:2024 is the international standard derived from it.
Why do high performance HMIs use grey?
A muted grey background with grey equipment lets colour carry meaning. When the normal state of the plant is grey, any colour on the screen is an abnormal condition or an alarm, and the eye goes to it straight away. On a screen where pumps, lines and vessels are all coloured, an alarm has to compete with everything else, and operators learn to ignore colour.
What are the display levels in a high performance HMI?
Most designs use four levels. Level 1 is an overview of the operator's whole span of control, watched all shift. Level 2 is unit control, where most routine operating happens. Level 3 is the detail of one piece of equipment or control loop, used for diagnosis. Level 4 is support, with interlock logic, diagnostics and procedures. Navigation runs down the hierarchy in a few clicks.
What is an analogue indicator on an HMI?
An analogue indicator shows a value as a pointer on a scale, with the normal operating range shaded and the alarm limits marked. When the pointer sits in the shaded band the value is normal and the indicator stays grey. When it moves out, the indicator changes to show the abnormal state. Operators can scan a row of indicators and see which values are out of range far faster than they can read and compare a row of numbers.
Key terms
Plain-language definitions from our glossary for the concepts this article leans on.
Standards and further reading
- ANSI/ISA-101.01-2015 Human machine interfaces for process automation systems, and the ISA-101 series (ISA)
- IEC 63303:2024 Human machine interfaces for process automation systems (IEC)
- IEC 62682:2022 Management of alarm systems for the process industries (IEC)
- EEMUA 201, Control rooms: a guide to their specification, design, commissioning and operation (EEMUA)
Related case studies and tools
- Real-time production analytics (case study)
Related reading
Working through something like this?
See how we approach these initiatives, or tell us what you are dealing with.