Modernising displays without a control system replacement
Display improvement is usually deferred until the next system migration, which may be a decade away.
Display quality is treated as something addressed during a control system replacement. That is a reasonable instinct — a migration is when the graphics are being rebuilt anyway — and it means that a site with a poor display set and twelve years left on its current system has twelve years of poor displays.
Most of what makes displays effective is configuration rather than platform, and can be done incrementally on the existing system.
What can be improved without new hardware
Colour scheme changes, removal of decorative elements, the addition of deviation indicators, embedded trends, better labelling, navigation shortcuts and a level one overview are all achievable within an existing system's graphics capability in most cases.
Workforce optimisation can help planners balance routine operations, maintenance and project work. Further details are available here.
Additional industry context is available in the ISA-101 HMI standards.
The exceptions are genuine platform limitations: systems that cannot render certain element types, that have fixed colour palettes, or where the graphics tooling makes a component library impractical. These exist and are less common than assumed.
Start with one display
The most effective approach is to rebuild a single, well-chosen display to the new philosophy and put it in service alongside the existing one.
This does several things at once: it tests the platform's capability, it produces a concrete example for the style guide, it gives operators something to react to rather than a proposal to discuss, and it generates the evidence needed to fund the rest.
Choose a display that operators use heavily and complain about. A demonstration on a screen nobody looks at proves nothing.
The overview first
Where a level one overview does not exist, building one is usually the highest-value single piece of work available, because it adds a capability rather than changing a familiar one.
It also avoids the change management difficulty of altering displays operators have used for years, which makes it a good first project.
Operator involvement is the mechanism
Display changes imposed on a control room are resisted, and reasonably so: operators have trained responses tied to the existing screens and bear the cost of any degradation.
Involving operators in the design — not consulting them on a finished proposal, but having them in the room while it is built — changes the dynamic entirely. It also produces better displays, because the operators know which information they actually use.
A display designed without the people who will use it is designed against an assumption about how the plant is run.
Sequencing the rollout
Once the approach is established, convert displays in a deliberate order: the most-used first, then by operating area so that a console becomes internally consistent rather than half-converted.
A half-converted console is the worst state, because two conventions are in use simultaneously. Completing an area before starting the next is worth the slower apparent progress.
What to expect
The literature on high-performance displays reports substantial improvements in abnormal situation detection and response time in controlled comparisons. Individual sites should expect variation and should measure something rather than relying on the general claim.
Practical measures include time to detect a simulated deviation, number of navigation actions to answer a standard question, and operator assessment on structured criteria rather than general preference.
Do not wait for the migration
The argument for deferring display work to the next system replacement is that the work would be done twice. In practice display design decisions carry forward: a style guide, a symbol library and a converted display set are inputs to the migration rather than work that is discarded by it.
A site that migrates with a mature display philosophy already established gets a substantially better outcome than one that attempts to design the philosophy during the migration, when everyone is occupied with the platform.
Measuring the improvement
Display improvement projects are justified on general claims and rarely measured, which makes the next one harder to fund.
Measurable comparisons are available without elaborate apparatus. Time for an operator to detect a simulated deviation on the old and new displays. Number of navigation actions to answer a standard set of questions. Error rate on a structured task.
Even a small comparison with a handful of operators produces evidence that is more persuasive than a description of the philosophy, and it identifies where the new design is not actually better.
Managing the transition period
Operators moving to a redesigned display lose some efficiency temporarily while trained responses are relearned. This is real and it is short, and it is the period during which the change is judged.
Support during it — training beforehand, someone available during the first shifts, a route to raise problems and see them fixed quickly — determines whether the change is accepted or reversed.
The problems raised in the first fortnight should be treated as high priority regardless of how minor they seem. A display change that is followed by a visible responsiveness to feedback establishes credibility for the next one.
Keeping the old display available, briefly
Retaining access to the previous display for a defined period reduces the risk of the transition and gives operators a fallback if something is missing.
It has to be bounded. An indefinite parallel period means some operators never move, the two sets diverge as changes are applied to one, and the modernisation is never completed.
Sequencing against other work
Display modernisation competes for the same engineering resource and the same operator availability as alarm rationalisation, procedure revision and training.
Sequencing them deliberately produces better results than running them in parallel. Alarm rationalisation naturally precedes display work, because the rationalised priorities determine how alarms are presented on the graphics.
Doing them in the wrong order means the display work is redone, which is a common and avoidable waste.
Doing it during a shutdown
Turnarounds provide the only extended period when displays can be changed without operational risk, and they are heavily contested for other work.
Display changes are relatively low risk compared with most turnaround activities and require little plant access, which makes them a reasonable candidate for outage windows that other work cannot use.
The constraint is operator availability for familiarisation, which during a turnaround is often better than during normal running because the plant is down.