4.5 SPC & Dashboards

Statistical Process Control (SPC) is the operational discipline that converts raw machine data into useful predictive intelligence. By continuously charting the key performance indicators of the SMT line, SPC enables early, minor corrections to process drift before defects occur. Dashboards act as the visual interface for this control, providing real-time guidance to operators and engineers on the floor. A successful SPC system is characterized by clarity and its ability to guide effective action when a metric begins to shift.

SPC Fundamentals in Production

At its core, SPC involves monitoring key performance indicators to distinguish between common cause variation (inherent, expected process noise) and special cause variation (a sudden, specific problem like a clogged nozzle or a blocked sensor).

• Process Control: SPC is fundamentally about controlling the mean (µ) and minimizing the variation (σ) of critical parameters, such as Solder Paste Transfer Efficiency (TE).
• The Goal: The engineering aim is to maintain processes in a state of statistical control, assuring that the process is consistently capable (Cₚₖ ≥ 1.33) of meeting engineering specifications.
• Visual Control: Data must be presented clearly. Simple charts displayed prominently on overhead screens help operators read the process status at a single glance, much like reading a fuel gauge.

Key Process Metrics (By Station)

Effective SPC requires tracking a few vital parameters that accurately predict downstream quality, rather than monitoring every available data point.

Station	Key Metric	Indicator of…	Typical Response
Printer (SPI)	TE Cₚₖ by Feature Family	Predicts bridges, tombstones, and HIP risk.	A stencil cleaning cycle should be initiated or squeegee pressure softly adjusted.
	Wipes/Board Trend	Stencil/paste hygiene drift.	The automated cleaning interval must be adjusted or stencil nano-coating wear inspected.
PnP	Starvation Minutes/Hour	Feeder management practices on the floor.	An operator must be assigned to load a spare feeder; the splice alarm threshold should be reviewed.
	Miss/Retry Rate by Feeder	Challenging component tape, a worn nozzle, or vision instability.	The tape peel path must be reviewed or the specific feeder and component reel evaluated.
Reflow	TAL / Peak Mean	Solder wetting completeness and available thermal margin.	A very small belt speed adjustment or a precise zone temperature update should be considered.
	Cross-Board ∆T	Convection uniformity and overall thermal profile balance.	Blower speed must be checked or the specific soak zone setpoint adjusted.
Inspection	False Calls/Board (by RefDes)	Program tuning capability and manual verification effort.	Programmer review of specific area lighting or ROIs.

Dashboards and Alarm Tiers

Effective dashboards use clear visual alerts to guide a standardized response protocol. Tying deviations to documented actions creates a smooth operational playbook.

Visualization

• Control Charts: The I-MR chart is a standard tool for tracking continuous data like SPI TE and Reflow TAL/Peak. It clearly illustrates the process mean (µ) and variation (σ) over time.
• Traffic Lights: Clear Green/Yellow/Red status indicators readily communicate the health of each parameter to anyone observing the line.

Alarm Tiers

Organizing alarms into tiers helps prevent alert fatigue, ensuring that genuine issues receive prompt attention.

Tier	Condition	System Response	Management Action
Green	Metrics remain within 3σ.	Continuous flow is authorized.	No action required. Continue standard operation.
Yellow (Warning)	2 of 3 consecutive data points outside 2σ, or a distinct run of 6 points trending in one direction.	A warning is logged, an alert is sent to the supervisor, and the first correction is recommended (e.g., “Wipe Stencil Now”).	Line Supervisor oversees the suggested corrective action.
Red (Stop)	A data point exceeds the Specification Limit (USL/LSL) or a sustained variation of the Cₚₖ floor occurs.	The line must be halted; engineering review is recommended to resume.	A formal root cause analysis must be begun to ascertain the primary issue.

Closing the Loop (The Corrective Cycle)

The primary function of the SPC system is enabling the closed-loop feedback that drives continuous process improvement.

1. Detection: An alarm tier (Yellow or Red) is triggered by a measurable metric drift, for example, if SPI TE µ drops by 5 points.
2. Correction: The operator addresses the First Correction outlined in the playbook, such as marginally increasing squeegee speed. This is typically a simple, single-parameter adjustment.
3. Annotation: The action is logged against the exact time and product on the control chart, creating a reference point labeled “Process Change.” This logging supports later engineering audits of the process.
4. Verification: The system monitors the subsequent 3–5 panels to objectively verify that the correction successfully shifted the mean back toward the center without increasing variation.
5. Engineering Fix: If routine operator corrections do not stabilize the metric, the issue is elevated and classified as a Special Cause (e.g., a worn stencil or a faulty feeder). This calls for a permanent mechanical or software solution and an update to the Golden Recipe.

Final Checkout: SPC & Dashboards

Requirement	Control Point	Management Focus
Metrics	Specific predictive metrics (TE Cₚₖ, TAL/Peak, Miss Rate) must be selected and tracked.	Streamlines data; ensures the team focuses on critical control points.
Action Plan	Yellow and Red alarms must be tied to a documented First Correction playbook.	Promotes clarity and consistency during drift events.
Data Integrity	All machines must be synchronized to a master time base; recipe changes and operator actions must be logged on the charts.	Ensures data is trustworthy for root cause analysis.
Visualization	Simple control charts and clear status indicators must be utilized on dashboards.	Supports prompt and standardized operator response on the floor.