How to Build a Shop Floor [Production Dashboard](/service/manufacturing-dashboard-software) Production managers make dozens of decisions per shift — and most of them depend on information that arrives too late. A machine goes down at 9 AM and the supervisor finds out at 2 PM during a walkthrough. Parts fall behind schedule by lunch, but no one escalates until the shift-end report lands on a desk. By then, the window to fix anything has closed.

A shop floor production dashboard solves this by putting critical data — machine status, actual vs. target output, OEE, cycle time — on screen in real time, where the people who can act on it can actually see it.

The concept sounds simple. The execution isn't. Dashboards fail regularly because the wrong metrics get displayed, data sources remain disconnected, or a single screen tries to serve operators, supervisors, and managers simultaneously. According to the Manufacturing Leadership Council's 2024 Data Mastery survey, 70% of manufacturers still collect data manually — which means most shops are building dashboards on top of a data foundation that isn't ready.

This guide covers the full build process: what you need before you start, the five-step process, the metrics worth displaying, common dashboard types, and the mistakes that cause dashboards to get ignored within weeks of launch.

Key Takeaways

  • Define 5–8 role-specific KPIs before touching any dashboard tool
  • Data quality determines dashboard value — manual or batch-sync data can't deliver real-time visibility
  • Role-based views (operator, supervisor, manager) are non-negotiable for usability
  • The 5-second rule: any critical metric must be readable in a glance — large text, high contrast, minimal clutter
  • Treat the dashboard as a living tool with a designated owner, not a one-time build

How to Build a Shop Floor Production Dashboard

Step 1: Define the KPIs You Need to Track

Start with pain points, not metrics. Ask: what problems on the floor go unnoticed until it's too late?

Common answers include untracked downtime, unclear whether the shift is on pace, no visibility into scrap until end-of-day, or cycle times running long with no one catching it. Each pain point maps to a specific KPI:

Pain Point KPI
Shift falling behind schedule Actual vs. target production count
Machine stops going unnoticed Machine availability / downtime
Equipment underutilized OEE (Availability × Performance × Quality)
Parts running slower than planned Cycle time vs. standard
Scrap discovered too late First-pass yield / scrap rate

Shop floor pain point to KPI mapping comparison table infographic

Limit the initial dashboard to 5–8 KPIs. Research on working memory consistently finds that people process 3 to 5 meaningful items at once under normal conditions — beyond that, decision speed degrades. A dashboard with 20 metrics is harder to act on than one with six.

Role also determines which KPIs belong where:

  • Operators need station-level data: current job, cycle count, machine status
  • Supervisors need line-level data: actual vs. target, downtime alerts
  • Managers need plant-level data: OEE, WIP, throughput, schedule attainment

Don't try to answer all three roles on one screen.

Step 2: Identify and Connect Your Data Sources

A shop floor dashboard typically pulls from three sources:

  1. Machine controllers — cycle counts, spindle status, run/idle/alarm states (via MTConnect, OPC-UA, or protocols like Fanuc FOCAS)
  2. ERP systems — job orders, scheduled quantities, due dates, planned cycle times
  3. Operator input — downtime reason codes, job selection, quality flags

The critical question is whether your sources are connected and how fast they update.

A dashboard fed by manual data entry or end-of-shift ERP batch exports isn't a dashboard. It's a historical report with a live-looking interface. Teams figure this out quickly and stop trusting it.

Evaluate your data infrastructure honestly:

  • Can your machines output data digitally, or do they require hardware adapters?
  • Does your ERP expose job order data through an API or native connector?
  • Are operator inputs captured at the machine, or on paper and entered later?

The integration burden varies significantly by environment. Platforms like Harmoni are built specifically to sit between ERP systems, machines, and operators — connecting controls from Mazak, Haas, Fanuc, Heidenhain, Siemens, DMG MORI, and others to ERP platforms like Epicor, Infor, and JobBoss in a unified real-time view.

For shops with mixed machine ages or siloed systems, that integration layer is what makes a real-time dashboard achievable without a multi-year IT project.

Step 3: Choose Your Dashboard Structure and Layout

Role-based views aren't optional — they're what separates a useful dashboard from a cluttered wall display that everyone learns to ignore.

Design each view for its audience:

  • Operator dashboard: Current job, actual vs. target count, cycle time, machine status for one workcenter
  • Supervisor dashboard: Line-level actual vs. target, machine status summary, active downtime events
  • Manager dashboard: Plant-wide OEE, throughput, WIP, schedule attainment

Three-tier role-based dashboard view operator supervisor manager comparison infographic

Apply the 5-second rule. Any critical metric should be understood within five seconds of glancing at the screen. Grounded in HMI design principles (ISA-101) and cognitive load research, it's a practical usability test: if someone has to read carefully to understand what the screen is telling them, the layout needs work.

Practical layout requirements for shop floor environments:

  • Large text — readable from 10–15 feet away
  • High-contrast colors — avoid pastels or low-contrast combinations
  • Color-coded status — green (running), yellow (idle/warning), red (downtime/alarm)
  • Minimal elements — fewer items per screen, more actionable information per item

ISA-101 describes a four-level HMI hierarchy moving from plant overview down to diagnostic detail. That same structure applies here: build the overview first, then add drill-down access for detail.

Step 4: Build the Dashboard and Connect Live Data

Start with the high-level view, not the detail.

Build actual vs. target and machine status summary first — these two displays alone provide enough visibility to change shift outcomes. Once that's working with live data, add drill-down views: downtime breakdown by machine, cycle time by job, OEE component split.

The distinction that matters most: every dashboard needs both static goals and live actuals.

  • Static goals = manually entered targets (planned count per shift, standard cycle time)
  • Live actuals = data pulled from machines or ERP in real time (parts made, current cycle time, machine state)

Without both, the dashboard has no context. A count of 47 parts means nothing without knowing the target was 60.

Get something on the floor fast. A working dashboard with two metrics builds more team adoption momentum than a perfect dashboard that takes six months to configure. Operators and supervisors who see real data on day one will engage — and their feedback in the first two weeks will shape the final design better than any planning document.

Step 5: Deploy, Test, and Iterate

Physical deployment matters. Consider:

  • Wall-mounted displays at end-of-line positions for supervisor visibility
  • Machine-side screens or operator terminals for workcenter-level views
  • Touch-enabled displays where operators need to input downtime codes or job selections

Once it's live, the real design work begins. Watch how operators and supervisors actually use it in the first two to four weeks. Which metrics do they reference when making decisions? Which ones do they ignore? Remove what isn't driving action and add what's missing.

Harmoni, for example, deploys machine-side operator command centers that put job data, work instructions, and quality checksheets directly at each workcenter. The same physical infrastructure that feeds the dashboard serves as the operator's primary interface — which reduces adoption friction significantly.

The dashboard is not a finished product. Processes change, targets shift, equipment gets added or replaced. Assign a specific owner — an engineer, operations manager, or IT liaison — who is responsible for keeping layouts and data connections current. Dashboards without owners drift out of relevance within months.

What You Need Before Building Your Dashboard

Preparation determines whether your dashboard becomes a decision-making tool or a screen that people walk past without looking.

Equipment and System Requirements

  • Machine connectivity: Machines need to output data digitally — via MTConnect, OPC-UA, Fanuc FOCAS, or similar — or have a hardware adapter in place. Harmoni, for instance, supports machines that have no digital output by using analog and current sensors that detect machine state from electrical lines, which means even legacy equipment without network capability can contribute live data.
  • ERP integration: Your ERP must expose job order data, scheduled quantities, and due dates in a consumable format — via API, connector, or native integration. If your ERP only allows batch exports, plan for that limitation upfront.

Inputs, Data, and Role Definitions

Define what gets collected automatically versus what requires operator input before you design any layout:

  • Automatic: Machine cycle counts, spindle status, run/idle/alarm states, job completion signals
  • Operator-entered: Downtime reason codes, job selection (if not RFID-automated), quality inspection flags

Skipping this step leads to dashboards that show gaps wherever operator input was assumed but not designed for.

Confirm role definitions before layout begins. A single view built for all roles typically serves none of them — operators, supervisors, and engineers each need different data at different levels of detail.

Skill and Operational Readiness

Assess who will maintain the dashboard after go-live. A dashboard that requires an IT ticket to update a target number will fall out of date within weeks. Operations or engineering staff should be able to adjust layouts, update targets, and modify data connections without developer involvement — or the dashboard will drift from reality faster than it can be corrected.

Key Metrics to Display on Your Shop Floor Dashboard

The metrics that belong on a shop floor dashboard are the ones teams can act on during the shift — not metrics that explain what happened after production ends.

Actual vs. Target Production Count

This is the most immediate shift-health indicator. Is the line ahead, behind, or on pace?

Displayed prominently in real time, actual vs. target gives operators and supervisors the information they need to self-correct before the gap becomes unrecoverable. Without it, the only signal a shift is behind is a supervisor's gut feeling or a late-day walkthrough.

Machine Status and Availability

Color-coded machine status — running, idle, in downtime — gives supervisors an instant picture of where production has stopped. Knowing a machine is down matters less than knowing immediately — before an hour of capacity disappears.

A 2023 Siemens/Senseye survey found the average facility experiences 20 unplanned downtime incidents and 25 lost production hours per month. Real-time visibility doesn't prevent every incident, but it compresses the time between a machine stopping and someone responding.

OEE (Overall Equipment Effectiveness)

OEE = Availability × Performance × Quality. Each factor exposes a different class of loss:

  • Availability: How much of planned production time the machine actually ran
  • Performance: How fast it ran compared to ideal cycle time
  • Quality: What percentage of parts were good on the first pass

OEE overall equipment effectiveness three-component availability performance quality breakdown

OEE.com benchmarks set the benchmark for OEE at 85%, with typical discrete manufacturers closer to 60%.

Displaying OEE live — rather than calculating it after the fact — lets teams spot degradation within the shift and respond before the day is lost.

Cycle Time vs. Standard

Comparing actual cycle time per part against the planned standard reveals whether a machine or operator is running slower than expected, and by how much. Persistent overruns often point to tooling wear, incorrect feeds and speeds, or program issues — each fixable during the shift rather than at job closeout.

Quality and Scrap Rate

Cycle time overruns and rising scrap rates often share a root cause — which is why quality belongs on the same dashboard. Tracking these three metrics alongside throughput keeps quality visible in context:

  • Parts produced vs. target
  • Parts rejected and rejection reason
  • First-pass yield trending across the shift

Early visibility into a rising scrap rate lets operators stop and adjust before a batch is lost. Harmoni captures this through a combination of real-time machine monitoring and operator-entered digital quality checksheets at each workcenter — so quality data is available at the dashboard without waiting for a separate quality system to be updated.

Types of Shop Floor Production Dashboards

Most manufacturing environments need multiple dashboard views, not one universal screen.

Overview / Control Tower Dashboard

A plant-wide view showing all machine statuses, aggregate OEE, and overall throughput. Used by plant managers and supervisors monitoring the whole floor. Typically displayed on large screens in central locations or accessible from any device.

Production / Operator Dashboard

A station-level view for a specific workcenter designed for the operator to self-monitor and act without waiting for supervisor intervention. It surfaces what operators need at the machine:

A station-level view for a specific workcenter, designed for the operator to self-monitor and act without waiting for supervisor intervention. It surfaces what operators need at the machine:

  • Current job and work order context
  • Actual vs. target part count
  • Cycle time and machine status

Harmoni's machine-side operator terminals put production data and job context in one place, so operators never have to leave the machine to find information.

OEE / Performance Dashboard

A deeper breakdown of availability, performance, and quality losses used by process engineers and continuous improvement teams. The focus is identifying systemic losses and prioritizing improvement projects, not moment-to-moment intervention.

Common Mistakes When Building a Shop Floor Dashboard

Displaying Too Much Information

Every metric added to the primary display competes for attention with every other metric. The 5-second rule breaks down when operators must scan 20 data points to find the one that requires action. Push diagnostic detail into drill-down views and keep the primary display focused on the 5–8 metrics that drive decisions.

Using Disconnected or Delayed Data

A dashboard fed by spreadsheet exports or manual entries is a report dressed up to look real-time. Teams recognize this quickly and stop trusting it. Without live connections to machines and ERP, the dashboard can't support decisions — it can only document what already happened.

Building One View for All Roles

Operators and managers need fundamentally different information. A single dashboard trying to serve both ends up serving neither. Build role-based views from the start, or use drill-down navigation to separate perspectives:

  • Operators need cycle counts and machine status for their workcenter
  • Supervisors need line-level throughput and queue visibility
  • Plant managers need OEE, on-time delivery, and capacity utilization across the floor

Role-based shop floor dashboard data needs by operator supervisor and plant manager

Treating the Dashboard as a Finished Product

A dashboard without a designated owner becomes outdated within months. Targets change, machines are replaced, processes evolve. Someone needs to own the dashboard — reviewing it quarterly, updating targets, and adjusting layouts as the operation changes. Without that ownership, the dashboard stops reflecting reality — and operators stop using it as a result.

Frequently Asked Questions

What is an OEE dashboard and how is it used on the shop floor?

An OEE dashboard displays Overall Equipment Effectiveness — Availability × Performance × Quality — live so operators and supervisors can see how effectively equipment is being used right now, not after the shift ends. That real-time view lets teams identify and respond to losses before the day's production goals are missed.

How can a shop floor production dashboard increase productivity?

Operators self-correct faster, supervisors intervene sooner, and downtime events get resolved with full context — because actual vs. target performance is visible the moment something goes wrong. The lag between a problem occurring and someone addressing it disappears.

What types of dashboards are used for shop floor production?

The most common types are production dashboards (actual vs. target output), OEE dashboards (availability, performance, quality), maintenance dashboards (machine health and downtime), and quality dashboards (scrap rate, first-pass yield). Each is typically designed for a specific role and decision type.

What is the 5-second rule for a shop floor production dashboard?

The 5-second rule is a usability principle stating that any critical metric on a shop floor dashboard should be understood within five seconds of a glance. It drives design decisions around large text, high-contrast color coding, and limiting the number of visible metrics per screen.

What data sources feed a shop floor production dashboard?

Shop floor dashboards pull from three sources:

  • Machine controllers — cycle counts, status, cycle time
  • ERP systems — job orders, scheduled quantities, due dates
  • Operator input — downtime reason codes, job selection, quality flags

How often should a shop floor dashboard be updated?

An effective shop floor dashboard should update in near real time — within seconds of a machine event or operator action. Dashboards that refresh on a delay of minutes or longer become reports. Teams stop trusting them, and the decision-making value disappears with that trust.