Explainability of AI models refers to the ability to present data-driven decisions in a transparent, understandable, and technically justifiable way. In industrial inline measurement technology, this is a key requirement for acceptance, auditability, and long-term process stability.
Thickness Measurement User Interface
Production processes are subject to quality standards, documentation requirements, and internal validation procedures. Decisions that influence scrap, rework, or process approvals must be technically justified. Black-box models without transparent decision logic pose risks for quality assurance and production release.
• Clear definition of input features (feature transparency)
• Traceable model structure or explainable decision logic
• Reproducible results for identical inputs
• Validatability against reference data
• Ability to re-validate when process conditions change
• Documentation of model limits and application scope
In 3D or 2D measurement systems, every AI decision is based on previously extracted geometric or optical features. The quality of the decision therefore strongly depends on structured feature definition. An explainable system shows which features were weighted, how they were weighted, and how they contribute to the final quality decision.
An AI model must not only work within the training dataset, but must also remain stable under real production conditions with process variation, material scatter, and environmental influences. Serial validation includes comparative measurements, boundary-case analyses, and statistical stability checks.
Explainable AI does not mean abandoning data-driven methods, but integrating them in a controlled way into industrial decision logic. In many cases, a combination of deterministic rules and data-driven models is beneficial in order to ensure both transparency and performance.
QuellTech integrates data-driven evaluation methods into 3D laser-based inline systems with the goal of assessing process-relevant features in a traceable way. The focus is on industrial robustness, clearly defined feature logic, and auditable decision structures—not on opaque black-box optimizations.
Inline measurement technology and industrial system integration:
Optical Inline Measurement Technology
3D laser technology as the basis for structured feature extraction:
3D Laser Scanner Measurement
We are happy to help you with that!
Stefan Ringwald
Technical Contact
If you have any questions about Optical Inline Measurement or would like a consultation from QuellTech on this subject, we will be happy to help.
We would like to help you to precisely evaluate your specific measurement task. Through an initial free test measurement of your application, we can give you an early assessment of the feasibility.
There is always potential for improvement, we will help you with that. Contact us for more information or to make an appointment for a consultation.
Would you like to assess the feasibility of your measurement task as early as possible? Contact us and take advantage of our free initial sample measurement.
By submitting the contact form, you agree that your data will be used to process your request (further information and instructions on withdrawal can be found in the privacy policy).
Your technically competent contact, Stefan Ringwald, will be happy to help you.
Our aim is to fully understand your specific technical requirements for your current measurement task.
Thanks to many years of experience with complex tasks in the field of 3D laser measurement, you will receive well-founded solution proposals from us.
Hinweis für Deutsche Unternehmen:
Für Investitionen in Ausrüstungen gilt seit Juli 2025 eine degressive Sonderabschreibung von bis zu 30% p.a. – befristet bis Ende 2027.
Je früher die Anschaffung erfolgt, desto länger kann dieser Vorteil genutzt werden.
(Dies ist keine Steuerberatung, bitte wenden Sie sich für Details an Ihre Steuerberatung.)
You are currently viewing a placeholder content from Vimeo. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
More InformationYou are currently viewing a placeholder content from YouTube. To access the actual content, click the button below. Please note that doing so will share data with third-party providers.
More InformationYou need to load content from reCAPTCHA to submit the form. Please note that doing so will share data with third-party providers.
More Information