Dashboards and Navigation

Report folders and dashboards#

Charts live in report folders; dashboards reference charts and place them on a grid.

That separation is deliberate and useful: one chart can appear on several dashboards. A plant-wide OEE chart can be on the executive summary, the plant overview, and the area detail page, configured once.

It has a corollary worth remembering: editing a chart changes it everywhere it appears. Before adjusting an axis to suit one dashboard, check where else that chart is used. Where two dashboards genuinely need different treatments, that is two charts.

The 12-column grid#

Panels are placed on a 12-column grid, with rows growing downward. Each panel has a left position, a top position, a width, and a height.

Practical sizes:

Twelve columns divides cleanly by 2, 3, 4, and 6, which is why four widgets across a row (3 each) or three charts across (4 each) both work. Five of anything across a row does not, and forcing it produces a ragged layout.

Building a dashboard#

1. Build the charts first, in a report folder. Charts are bound to intervals, so getting them right before placing them saves rework.
2. Create the dashboard.
3. Add a panel per chart, with its position and size.
4. Add a Text panel as a heading.
5. Add a Navigation panel if this dashboard is part of a set.

From the CLI, each panel is one command specifying the chart and its grid rectangle — which makes a consistent set of area dashboards straightforward to generate rather than place by hand.

Navigation turns dashboards into a system#

A single dashboard is a page. A linked set is a system someone can actually use to investigate something.

The pattern that works is summary to detail:

Plant Overview            < the one on the wall
+-- Utilities             < area detail
|   +-- Boiler House
|   +-- Compressed Air
+-- Production
|   +-- Line 1
|   +-- Line 2
+-- Quality

Each level shows what matters at that level and links down. A manager sees a red indicator on the overview, clicks into the area, clicks into the line, and finds the cause — without navigating a model tree or knowing which chart to look for.

Use the Navigation chart type for the links. Two habits make a set navigable:

Link down and back up. Every detail dashboard should offer a way back to its parent. Users who reach a dashboard from a link and then cannot get out will use the browser back button and eventually stop exploring.

Keep the structure shallow. Three levels is comfortable. Five means nobody finds anything twice.

Dashboard loops#

A dashboard loop cycles automatically through a set of dashboards on a timer. This is what puts Infohub on a control room wall or a factory floor screen with no one operating it.

Loops are a distinct capability, configured against a report server template with the dashboards to include and the timing.

Design for the medium, because signage is not a desktop:

• Far fewer elements per dashboard. Read from four metres, a screen holds three or four things, not twelve.
• Much larger type and elements. Widgets over tables; a table is unreadable at distance.
• Long enough dwell time. Eight to fifteen seconds per dashboard. Faster and nobody finishes reading; slower and the loop feels stuck.
• No interaction assumed. Nobody will hover for a tooltip or click through. Everything needed must be visible.
• Colour carrying the message. At distance, colour is read before numbers. Limits doing their job matters more here than anywhere.

Signing in, and the user profile#

Every Flow Server page carries a user profile icon in the top right, and it doubles as the sign-in indicator:

• A silhouette — nobody is signed in.
• Initials — those of whoever is.

Clicking the silhouette opens the sign-in dialog, where a Windows account and password are entered.

That icon answers a support question that comes up constantly: "why can't I see the dashboard?" A viewer looking at a silhouette is anonymous, and signing in is what makes the dashboards, reports and forms they have access to appear. Nothing announces that content is being withheld — the page simply shows less.

So when somebody reports a missing dashboard, have them check the top-right corner before anyone starts looking at permissions. See Users, Groups and Access.

Scope and audience#

A dashboard should have one audience and one purpose. The most common failure is a single dashboard trying to serve a control room, a plant manager, and a board.

Building four dashboards is less work than the arguments produced by one dashboard that satisfies nobody.

Consistency across a set#

When there are twenty dashboards, consistency is what makes them usable.

• Same layout convention. If widgets are along the top on one area dashboard, they are along the top on all of them.
• Same colour meanings everywhere, driven by limits and enumerations rather than set per chart.
• Same naming. Line 1 Overview, Line 2 Overview — not Line 1 Overview and Line Two Dashboard.
• Same period. Two panels on one dashboard showing different periods without saying so is a reliable source of misreading.

Where the same dashboard shape repeats per area, chart templates keep them aligned as they evolve.

Diagnosing#

See also#

• Dashboard Layout Principles
• How Infohub Charts Work
• Exporting and Scheduling Reports
• Chart Templates and Releases