EDGEBIC Dashboards Explained: Four Tabs for Three Audiences

The EDGEBIC by User Solutions Dashboard Cockpit is one screen with four tabs, built for three different audiences, and every number on it carries a small chip that tells you how much to trust it. Where a report answers one question deeply, the dashboard answers "how are we doing right now?" at a glance. This guide covers what each tab is for, who should be reading it, and the two design decisions (data-quality chips and manual refresh) that shape how you use the whole thing.

The dashboard is the layer above the report catalog: tiles tell you something is off, reports tell you which job and which hour. For the click-by-click version of everything below, see how to use the dashboard tabs.

Four Tabs, Three Audiences

Nothing is duplicated for the sake of it. Each tab exists because a different person asks a different question of the same database, and giving all four to everyone is how dashboards end up ignored.

The Chips: How Much to Trust Each Number

The design decision that makes this cockpit unusual is that every tile declares its own provenance.

That last row matters more than it sounds. The Safety and Quality tiles on the standup board sit at Offline because incident and defect logging are separate data sources; the exec margin tile sits at Offline until revenue is captured. A dash is honest absence. A zero would read as perfect performance, and a fabricated 100% would flatter every plant equally. The same principle runs through the reports, where OEE reads "n/a" instead of inventing a quality factor.

Hover any tile's info icon and you get its exact formula and data source in a sentence. That is the difference between a tile you can defend in a meeting and one you have to caveat.

The Capacity Tab: The Working Default

You land here. Six hero tiles run across the top:

Below them, a work center utilization chart plots assigned against available hours per station, capacity trends plot total, assigned, and available over time, and an alerts panel lists current warnings. Sub-tabs go deeper: a capacity detail grid per work center with a department summary strip, a jobs view splitting scheduled from at-risk jobs, and a shifts view.

The Job Status Donut

Beside the active jobs list sits the donut, and it is worth learning properly because it encodes a distinction most shops make informally.

Two subtleties. Whether a job has physically started is a small badge on the active jobs list, not a slice, because starting says nothing about whether it will be on time. And classification runs against the delivery-ready date, which is the scheduled end plus the product's end-item lead time, which is why job rows show a "Ready" date next to the "Due" date. The list sorts worst first, so the donut and the list together are your attention queue.

The Standup Tab: The Morning Huddle

Switch to Standup and the view changes to a dark theme built for a wall screen. It reads top to bottom in the order a huddle actually runs.

The SQDC band covers safety, quality, delivery, and cost, the four pillars of a lean daily management board (the framework is standard practice in the APICS and ASCM body of knowledge most planners already share). Delivery carries on-time in full; cost carries actual against estimated cost variance; safety and quality sit offline until their logs exist.

The andon strip gives one tile per work center, each showing that machine's andon state, and its colors are inferred from schedule actuals rather than machine sensors:

That inference has one requirement: operators must punch in at operation start. A board that is all red by mid-morning is telling you about punch discipline, not about machines.

Below the strip, the bottleneck card and the late-job punch list, which on a good day simply says all orders are on track.

The Exec Tab: Ninety Days and Multiple Plants

Four strategic tiles cover throughput over the last seven days, on-time in full over the trend window, margin (offline until revenue exists), and backlog, which is a Live number because open order counts are persisted facts independent of any schedule.

The multi-plant roll-up assigns each job to one plant using a dominant-department rule: the department whose work centers consumed the most scheduled hours for that job wins it. That is a clean rule with one prerequisite, which is that every work center belongs to a department. A department whose throughput looks impossibly low is usually one with unassigned work centers rather than a quiet plant.

Top and bottom products rank the last 90 days by volume and by worst on-time performance, so the chronically late products that also ship in quantity surface first. That pairing is the list worth taking to an engineering review.

The Planner Tab: Measuring the Plan Itself

This is the tab that measures the schedule rather than the shop.

Schedule adherence is the share of operations that started within an hour either side of plan over the past 14 days, with a target of 90% or better. Its delta note ("35 ops moved off plan") is an investigation count rather than a statistic. Our schedule adherence KPI guide covers the generic metric; the number here is Live, because it is computed from real start events.

Attainment is the share of completed jobs that hit their due date, target 85% or better.

The 14-day heatmap colors every work center and day cell by load, which is assigned hours divided by that day's capacity. Red means oversold: more hours booked than the day holds. Hover a cell for the raw numbers. What to do about a red cell is the subject of reading a red day.

The Resource Calendar: The Planner's Ledger

Alongside the cockpit, the Resource Calendar in the Schedule area is the capacity ledger: work centers by time period, switchable between total capacity, hours assigned, backlog, available capacity, and utilization, in day, week, month, or quarter buckets.

Backlog is the mode most shops under-use. Assigned answers "what hours land in this period?"; backlog answers "how much work is still ahead of this station as of this period?". It is a forward sum of remaining hours, so it burns down as work is consumed: a 24-hour step spread over three days reads 24, then 16, then 8, then 0 rather than repeating 24 until it finishes. A backlog that grows week over week means work is arriving faster than the station clears it, which is a lateness warning weeks before the donut turns red.

Click any non-zero assigned or backlog cell and a job detail popup lists the exact allocations behind the number, with job, product, work center, shift, date, instance, hours, and pieces. The rows sum back to the cell under the same filters, which is what makes the number arguable rather than mysterious.

Dashboard or Report?

Same database, different lenses. Dashboards are glanceable and read the plan at a summary altitude. Reports are the drill-down and carry the authoritative formulas plus the built-in column glossary. For capacity specifically, the Work Center Utilization report is the authority, and the department capacity view shows the same numbers rolled up. When the two seem to disagree, they are usually answering different windows: a heatmap cell is one day, a report averages a range.

For the ways a correct tile still gets read wrong, see dashboard misreading mistakes. A wider cross-plant view built on these same signals is described in the EDGEBIC Control Tower roadmap, which is explicit that it is a planned direction and not a screen you can open today. For the whole system, the complete EDGEBIC guide is the pillar, and the documented results show what shops do with the numbers.

Want to see your own plant on this cockpit? Bring an export to a demo.