DeepDivers

This is a case study from Wilson Prasad’s experience within the industry he is sharing to drive a learning exercise with his defined quote.

Consistency in quality is not achieved by a single control point, but by harmonizing materials, process, and people into one reproducible system. – Wilson Prasad, 2026

The Problem

On a high-volume toothpaste manufacturing line, several consecutive batches exhibited low viscosity immediately after manufacture.

However, after 24 hours at rest, the viscosity moved comfortably into specification.

This created a critical operational dilemma:

• Release immediately and risk OOS results
• Delay release and disrupt production flow
• Pause manufacturing until root cause was understood

The real challenge was not just meeting specification — it was achieving right-first-time viscosity at Day 0, without relying on time-dependent stabilization.

Define

Historical batch data was reviewed to identify patterns.

Out-of-trend viscosity results were correlated against:

• Raw material lots (particularly silica grades)
• Equipment configuration changes
• Production timeframes
• Operator shifts

This established that the issue was systemic… not random.

Measure

A structured rheology study was designed:

• Viscosity measured at 0, 24, 48, and 72 hours
• Silica characterized for:
  • Particle size distribution
  • Surface area and flow properties
  • Moisture content
• Mixing practices documented directly on the manufacturing floor
• Shear rates and mixing energy reviewed against SOP targets

The goal was to move beyond a single-point QC result and understand viscosity development over time.

Analyse

The investigation revealed a classic structured-fluid behavior:

Toothpaste viscosity is governed by a time-dependent silica polymer network formation.

The hydrocolloid system (e.g., CMC or similar structuring polymers) interacts with hydrated silica particles to form a three-dimensional rheological network.

Key findings:

• Small shifts in silica texture and particle-size distribution altered hydration kinetics
• Variations in shear energy during mixing influenced network formation
• Minor moisture variation changed silica wetting efficiency
• Polymer activation time varied depending on mixing sequence

The result:

Day 0, Viscosity could be low even when final composition was correct.

The system simply had not yet completed its structural build.

This explained the 24-hour recovery.

Improve

The solution required harmonizing materials and process — not adjusting one variable.

Actions implemented:

• Tightened silica acceptance criteria (PSD, moisture, flow)
• Standardized mixing speed, shear profile, and hydration hold time
• Optimized structuring agent grade and ratio
• Locked mixing sequence to ensure proper polymer activation
• Reduced uncontrolled shear variability between operators

Control

Sustainability measures included:

• Updated work instructions and visual SOP guides
• Operator retraining on critical rheology drivers
• Routine viscometer calibration checks
• Temporary multi-timepoint viscosity monitoring until stability was proven
• Ongoing trending of silica lot performance

Once stabilized, Day 0 viscosity consistently landed within target range.

Production resumed without reliance on delayed stabilization.

Key Technical Insight

Viscosity was not a laboratory number.

It was the emergent property of a dynamic silica–polymer network forming under specific shear and rest conditions.

In structured fluids like toothpaste:

• Rheology is time dependent
• Thixotropy must be understood, not ignored
• Small raw material variation can amplify under shear

This was not a QC failure.

It was a systems interaction problem.

Takeaways:

1️⃣ Rheology Is a System Property

Viscosity sits at the intersection of raw material behavior, equipment capability, and operator execution.

2️⃣ Time Matters

For thixotropic systems, Day 0 and Day 1 results may reflect kinetic differences — not compositional failure.

3️⃣ DMAIC Still Works

Define–Measure–Analyse–Improve–Control remains one of the most effective frameworks for solving complex manufacturing variability — even in mature, high-volume processes.

---

Authors Note: Wilson Prasad also known as user name muefatiaki1966 is trying to leverage his extensive experience within the industry to educate and invoke discussion for topics of interest.