GUIDEBOOK · CEA-4 APP TRAINING

Canvas Builder

Canvas Builder

Canvas Builder is a side-by-side comparison workspace. You choose what the columns mean — different scenarios, different what-if variants, or different years of a pathway — then place cards (plots, KPIs, maps, notes) that are drawn for every column at once.

It is the answer to “I have run the same analysis four ways; now show me the difference.”

When to use it

  • Comparing design alternatives held in sibling scenarios
  • Comparing supply what-ifs within one scenario
  • Reading a pathway year by year
  • Assembling a small report page for a meeting

For a single scenario’s results, the ordinary Visualisation plots are quicker. Canvas earns its keep the moment you need two or more things beside each other.

The four comparison modes

ModeColumns are…Typical question
Inter-scenarioSibling scenarios in the project”Which masterplan performs best?”
Inter-what-ifWhat-if variants under one scenario”Gas boiler vs heat pump?”
Pathway (single)State years of one pathway”How does the district evolve to 2050?”
Pathway (multi)Several pathways”Business-as-usual vs accelerated retrofit”

The mode is chosen when you create the view and decides what a column can point at. Cards are shared across columns, so adding one plot adds it everywhere — that is what keeps the comparison honest.

Card types

CardShows
PlotAny CEA plot, drawn per column with that column’s data
KPIA strip of headline numbers
MapA geographic layer
TextFree-form notes, per column
DividerA visual separator, mirrored across columns

Cards are placed on a free-form grid, so you can drag and resize them into a layout that reads the way you want.

KPIs

Canvas draws on a KPI registry covering nine domains:

DomainKPIsDomainKPIs
Architecture4Heat rejection3
Costs3Networks8
Demand3Optimisation2
Emissions2Solar4
Final energy5

A KPI is computed on demand from the results already on disk. If the feature behind it has not been run for that column, the KPI reports as unavailable rather than showing a misleading zero.

Comparing fairly: y-axis alignment

By default each plot scales to its own data, which makes two columns look similar even when their magnitudes differ by an order of magnitude. Canvas can align the y-axis across columns that share a slot, so the bars are drawn to one common scale.

Turn this on before drawing conclusions from bar heights. It is the single most common source of misread comparisons in a side-by-side layout.

Workflow

  1. Open Canvas from the app navigation.
  2. Create a view and pick one of the four comparison modes.
  3. Choose what each column points at — scenarios, what-ifs, or pathway years.
  4. Add cards. Each is drawn for every column.
  5. Enable y-axis alignment where you are comparing magnitudes.
  6. Arrange, resize, and annotate with text cards.

Your first canvas

A run-through from an empty workspace to a readable comparison. The example reads a pathway year by year, but the shape is the same for all four modes.

Before you start

Canvas reads results; it never runs anything. Whatever you want to compare has to have been run already — for a pathway that means baked and simulated states, for sibling scenarios it means each scenario has the features you plan to plot.

An empty column almost always means the analysis behind the card has not been run for that column, not that Canvas failed.

1. Open Canvas and create a view

Open Canvas from the app navigation and create a view. You are asked for a comparison mode straight away, because it decides what a column is allowed to point at:

pick thiswhen your columns are
Inter-scenariosibling scenarios in the project
Inter-what-ifwhat-if variants under one scenario
Pathway (single)the state years of one pathway
Pathway (multi)several pathways

For the pathway example, choose Pathway (single).

The mode decides what a column is allowed to point at, but it is not a one-way door: you can come back and change what you are comparing — adding or removing scenarios, or starting a different comparison — without rebuilding the canvas from scratch.

2. Point the columns at something

Each column gets its own target: a scenario, a what-if, or a pathway year. For a pathway view, add one column per state year you want to see — 2030, 2040, 2050.

Column order is yours to set, and for a pathway it is worth putting them in chronological order so the eye reads left to right as time.

3. Add cards

A card is added once and drawn for every column. That is the mechanism that keeps a comparison honest: you cannot accidentally plot different things in different columns.

carduse it for
Plotany CEA plot, drawn per column with that column’s data
KPIa strip of headline numbers
Mapa geographic layer
Textyour own notes, written per column
Dividera visual break, mirrored across columns

Start with one plot and one KPI strip. Cards sit on a free grid, so drag and resize them once you can see what you have.

4. Turn on y-axis alignment

Do this before you read anything off the bars. By default each plot scales to its own data, so two columns an order of magnitude apart can look nearly identical. Aligning the y-axis across columns that share a slot draws them to one scale.

This is the single most common way a side-by-side layout misleads.

5. Annotate and keep it

Add Text cards for what the reader needs to know — why these years, what changed between them. Text is per column, so it can carry a note specific to one case.

Canvas autosaves as you work, and reopens the last canvas you had open for that project and scenario — so you can leave and come back without losing the layout.


Common issues

“A column is empty.” The feature behind the card has not been run for that column’s scenario, what-if, or pathway year. Canvas reads existing results; it does not run analyses.

“A KPI shows as unavailable.” Same cause — the underlying results are missing. This is deliberate: an absent result is reported, never substituted with zero.

“Two columns look identical but the numbers differ.” Y-axis alignment is off, so each plot is using its own scale. Turn it on.

“What-if columns are missing.” What-ifs come from the parent scenario. Run Final Energy first — it creates the what-if that the other LCA features and their KPIs read.


← Back: District Evolution Pathways | Back to Index

Source: view raw on GitHub ↗