FAQ – How Do I Perform a Correlation Study Between My Laboratory Instrument and SpectraTrend HT?

Answer

There is no universal correlation template that applies to every SpectraTrend® HT installation. Instead, HunterLab recommends performing an application-specific correlation study using paired laboratory and inline measurements collected under actual production conditions.

Why Is a Correlation Study Necessary?

STHT and laboratory instruments measure products differently due to:

• Different geometries.
• Different optical designs.
• Different illumination conditions.
• Different measurement environments.
• Different product temperatures.
• Different surface characteristics.

As a result, relationships between the systems are application-specific.

What Is the Best Solution?

The best approach is to establish a structured correlation methodology using actual production data. The objective should be to determine:

• How STHT responds to process changes.
• How laboratory measurements respond to the same changes.
• Whether meaningful relationships exist between the two systems.
• How STHT can be used to improve process control.

Best Practices

• Collect measurements from both systems on the same production lots.
• Include acceptable and marginal product conditions.
• Record process variables during testing.
• Evaluate trends rather than individual measurements.
• Collect sufficient data to capture normal process variation.
• Repeat studies when products or processes change significantly.

Practical Guidance

A successful correlation study is typically performed using the following process:

Step 1 – Define the Objective

Determine what you are trying to achieve. Examples include:

• Matching pass/fail decisions.
• Detecting process shifts.
• Understanding product variation.
• Establishing process control limits.
• Predicting laboratory measurements from inline measurements.

The objective should focus on business outcomes rather than simply making the numbers match.

Step 2 – Collect Paired Data

Collect measurements from both systems using the same production lots. For each sample, record:

• STHT measurements.
• Laboratory measurements.
• Product identification.
• Process conditions.
• Date and time.

The goal is to build a representative dataset covering normal production variation.

Step 3 – Evaluate Relationships

Review the collected data and determine whether meaningful relationships exist. Questions to consider include:

• Do both systems trend in the same direction?
• Is there a consistent offset?
• Is the relationship linear?
• Are certain process conditions influencing the relationship?
• Are both systems identifying the same product changes?

In many applications, trend agreement is more important than numerical agreement.

Step 4 – Determine an Operating Strategy

Use the results to establish how STHT will support the process. Examples include:

• Establishing warning and action limits.
• Identifying process drift.
• Triggering laboratory verification.
• Monitoring product consistency.
• Supporting operator decision-making.

The objective is to use STHT to improve process control rather than replace laboratory measurements.

Step 5 – Validate and Refine

Monitor the correlation over time and refine the strategy as needed. Consider:

• Product changes.
• Raw material changes.
• Process modifications.
• New customer requirements.
• Seasonal or environmental influences.

Correlation should be viewed as an ongoing process rather than a one-time exercise.

Key Takeaway

HunterLab does not provide a universal correlation template because every installation is unique. The most successful correlations are built from actual production data and focus on understanding process behavior rather than creating mathematical conversions.

Do you need more information? Submit a ticket and a support team member will reach out to you soon!

To learn more about Color and Color Science in industrial QC applications, click here: Fundamentals of Color and Appearance