1. Program Setup
Define labs, tests, sample handling, data fields, schedule, confidentiality rules, and decision thresholds.
KAT Quality Loop
A weekly asphalt laboratory QA program that turns test variation into improvement.
Multiple laboratories test the same or equivalent asphalt sample on a repeated schedule. Kriz Asphalt Technology analyzes the results, identifies likely sources of variation, and helps labs improve the data quality that production, acceptance, and customer decisions depend on.
Built from Pavel Kriz's experience with asphalt laboratory QA programs, refinery-facing process improvement, Lean Six Sigma methods, and practical binder-test interpretation.
Equivalent sampleEach participating lab tests equivalent material so differences become visible.
Weekly rhythmRepeated testing creates trend visibility rather than one-time isolated snapshots.
Statistical diagnosisBias, variance, RMSE, repeatability, and reproducibility guide action.
Continuous improvementThe goal is better data quality, better production control, and better decisions.
Program model
Quality assurance as an improvement loop.
A strong lab QA program should help capable laboratories keep improving. It gives lab managers and leaders a clearer view of precision, trend behavior, practical impact, and the action most likely to create value.
- Lab results disagree enough to affect production or release decisions.
- Different operators or instruments appear to produce different answers.
- Borderline specification results create rework, delay, or customer concern.
- Management needs a simple view of lab quality without losing the technical truth.
- Existing proficiency testing does not translate into practical improvement.
How it works
From repeated testing to targeted action.
The program can be scoped for one company network, independent commercial labs, agency labs, producer labs, or a mixed stakeholder group where the shared goal is higher confidence in asphalt binder data.
2. Identical Sample Testing
Labs test the same or equivalent asphalt sample on a repeated weekly or agreed schedule under routine operating conditions.
3. Statistical Analysis
Results are reviewed using RMSE, bias, variance, trends, repeatability, reproducibility, and practical test limits.
4. Evidence Review
Analysis separates likely instrument, operator, method, sample-preparation, process, and material signals.
5. CI Options
Recommendations focus on the highest-value actions: training, SOP review, troubleshooting, control charts, or deeper root-cause work.
6. Follow-Through
Progress is monitored over time so labs can see whether corrective action improved precision and decision confidence.
Diagnostic capability
Analysis that points to the likely source of variation.
The value is not only in knowing that a lab is different. The value is in understanding why the difference may exist and what action is worth taking.
Instrument Effects
Identify whether one instrument behaves differently from the lab average, the network average, or its own historical trend.
Operator Effects
Separate operator-to-operator differences from true material behavior and method precision limits.
Method And SOP Drift
Review whether conditioning, sample prep, loading, trimming, calibration, or reporting practice may be driving the difference.
Production And Release Signals
Use better data quality to support production control, release confidence, customer explanations, and fewer false alarms.
Operating philosophy
A precision system for capable laboratories.
The program is designed to make good laboratories stronger. Results are interpreted against method precision, practical impact, trend behavior, and the operating decisions the data is supposed to support.
Confidential by designLab identities, raw data, and corrective actions are governed by the agreed program rules.
Action-orientedReporting should lead to the next useful technical step, not just a ranking.
Technically groundedInterpretation uses asphalt test precision, rheology, and practical lab behavior.
Improvement-focusedTraining, visits, troubleshooting, and CI work are available when analysis suggests value.
Engagement options
A program that can start small and mature over time.
The first version can be a focused pilot with a few labs and a narrow test set. As value is proven, the same structure can expand to more labs, more tests, richer dashboards, and deeper CI support.
Pilot Loop
Limited lab group, focused test set, clear sample protocol, initial statistical reporting, and decision on whether to scale.
Managed QA Loop
Ongoing repeated testing, scheduled analysis, periodic summary reports, and practical corrective-action guidance.
Evidence Deep Dive
Focused investigation of a lab, operator, instrument, method, or recurring variation pattern.
Training And SOP Review
Targeted training, method review, SOP comments, and practical decision rules for lab teams.
On-Site Troubleshooting
Site visit support where direct observation of workflow, instrument behavior, or sample handling can change the outcome.
Network Leadership Support
Help for producers, agencies, or lab networks that need governance, communication, and technical stewardship.
Better data quality creates better production decisions.
KAT Quality Loop can begin as a pilot for a small lab group, a specific test method, or a priority quality objective. The first discussion should define the labs, tests, sample logistics, confidentiality rules, and the decisions the program needs to improve.