RCPs Are Not Outdated. The Workflow Built Around Them Is.
The reflected ceiling plan is still essential. The mistake is asking one drawing to carry design intent, photometry, controls, quantities, approvals and every revision that follows.
I do not think reflected ceiling plans are outdated. I think the industry's dependence on them is.
An RCP is an essential coordination drawing. It tells the team where the ceiling elements sit. We need it. But lighting design is not a ceiling-location exercise. The RCP shows where. It rarely explains why.
The RCP should remain an output of the lighting workflow. It should stop being mistaken for the whole workflow.
A 300 mm change the drawing cannot understand
Imagine a wallwasher placed for a 2.7-metre wall, a particular texture and a specific stand-off distance. Then the joinery moves 300 mm. The symbol, quantity and fixture code can all remain correct while the lighting effect becomes wrong.
The problem is not that the RCP failed. The problem is that nobody connected the symbol to the surface, intention and calculation that created it.
The seven layers of a lighting project
| Layer | What it carries | What happens when isolated |
|---|---|---|
| Space intelligence | Room use, geometry, materials, daylight, occupancy and constraints | Targets and fixture choices lose context |
| Design intent | Hierarchy, focal surfaces, atmosphere, tasks and scenes | Symbols remain, meaning disappears |
| Luminaire intelligence | Photometry, optics, drivers, colour, mounting and accessories | Substitutions become visual guesses |
| Calculation intelligence | Assumptions, grids, observers, results and alternatives | Reports cannot explain later changes |
| Control intelligence | Zones, scenes, timing, sensors and override | Controls are added after the design |
| Commercial intelligence | Schedule, BOQ, approvals, lead times and procurement | The priced project diverges from the designed project |
| Revision intelligence | What changed, why, who approved it and what else is affected | Every update creates another truth |
The traditional workflow is a collection of islands
The architect issues a plan. The lighting designer marks an RCP. A fixture schedule is created in Excel. Calculations are built in DIALux. Renders are stored in a presentation. Meeting decisions sit inside notes, emails or WhatsApp. Vendor quotations arrive as separate spreadsheets. Site changes appear in red marks on a PDF.
Every file can be individually correct and the project can still be collectively wrong.
The problem is not that professionals are careless. The problem is that the workflow asks people to manually preserve the same intent across too many disconnected places.
What an AI-based workflow should actually do
AI in lighting design should not begin by pretending it can replace taste, judgement or authorship. It should begin with the repetitive work that causes context to disappear.
A useful AI workflow can read a floor plan, identify rooms and zones, organise project information and help structure the first set of lighting requirements. It can connect a space type to task, atmosphere, target levels, likely fixture classes and control intent.
It can help create a fixture register that carries performance, mounting, optical, electrical and commercial information. It can then connect that register to the drawing and BOQ so a change does not have to be manually repeated across every file.
It can flag missing information, inconsistent quantities, incompatible controls, incomplete product data and obvious compliance risks. It can help turn meeting decisions into actions and revisions instead of leaving them inside a transcript.
Most importantly, it can preserve the chain from intent to execution.
The RCP becomes one view of the project
In a connected workflow, the RCP remains important, but it becomes one output from a richer project model.
Select a luminaire symbol and the team should be able to see what it is, why it was selected, which spaces use it, which photometric file is attached, which driver it requires, how it is controlled, how many are in the current revision and whether the product has been approved.
Change the fixture and the connected schedule, BOQ and review status should update. Change the room use and the project should flag that the existing light target, fixture logic or controls may no longer be appropriate.
That is the difference between digitising a drawing and building a lighting workflow.
Where human judgement remains essential
AI can organise options. It cannot decide what a space should mean to the people using it.
A lighting designer still decides where the hierarchy sits, which surfaces deserve attention, how darkness is used, how the architecture is revealed and how the space changes over time. The designer also carries responsibility for validating performance, interpreting standards and understanding site reality.
Technology should support design. Technology should not replace design.
The best AI workflow is therefore not the one that produces the most automatic output. It is the one that gives the designer more time to compare, test and refine meaningful options.
The lighting information model
The next workflow is not a smarter RCP. It is a lighting information model from which the RCP, fixture schedule, calculation brief, control narrative, BOQ and presentation can all be produced.
Move the joinery and the system should know which wallwash relationship is affected. Change the fixture and it should know which calculation, driver, control zone, BOQ line and approval record need attention.
The model Alya is building
I want the designer to remain the author. I want Alya to become the memory, translator and early-warning system across floor plans, photometry, calculations, fixture schedules, presentations, BOQs and project history.
AI should not draw a more impressive plan. It should stop the project from forgetting why the plan was drawn that way.
The RCP is not what becomes outdated. The workflow in which a static drawing is copied into five documents, revised manually, priced separately and defended from memory is what becomes outdated.
Primary Sources and Fact-Check Notes
- ANSI/IES LP-16-22 Reference for control intent beyond ceiling symbols.
- DIALux and open BIM Official BIM workflow information.
- ANSI/IES LP-1-24 Quality-lighting reference supporting a broader-than-plan view.
Fact-checked 25 August 2026. Confirm the adopted standard and local requirements for each project.