newQC did not emerge from a whiteboard. It emerged from eight research projects, three million euros of public funding, field deployments in retail operations, and consumer studies with 240 participants. This is what that journey looks like — and why it matters for buyers evaluating the platform today.
Key results
- Active in research projects since 2018
- 10 out of 12 projects were initiated and led by tsenso
- Strickly focused on 4 research tracks
- € 3,3m public reserach funding secured
- Out of the projects 10 lasting partnerships emerged
USE CASE – FROM RESEARCH TO PRODUCT
Why Research History Is a Buyer-Relevant Fact
Most quality management software is built by software companies. The domain knowledge is acquired, not developed. The models are generic, not validated against real produce under real operational conditions.
newQC is different in a specific and verifiable way: every core capability — the NIR scanner models, the shelf-life predictions, the supply chain data architecture, the data standards implementation — was developed and validated through publicly funded research programmes with named academic and industry partners.
That means two things for a buyer:
- The underlying science has been peer-reviewed and independently validated — not just tested internally
- The product development was funded by public institutions that required documented outcomes — the final reports are publicly available
Four Research Tracks, One Converging Product
The research was not a single linear programme. It ran across four parallel tracks that each solved a distinct problem — and converged into an integrated platform.
| Track 1 Meat Shelf-Life Prediction
Projects: FreshIndex, Zukunftslabor → Scientific shelf-life model for fresh meat. Validated dynamic freshness prediction replacing fixed best-before dates. |
| Track 2 Fruit Quality Assessment — Scanner & Inspection App
Projects: FreshAnalytics, FRED → NIR scanner development and calibration. Guided inspection workflow. Field validation at REWE Langel (June–August 2021). Consumer preference classifier validated with 850 samples, 240 participants. |
| Track 3 Supply Chain Quality Tracking & Prediction
Projects: FriDa, FreshCloud, FreshTwin, FRED → End-to-end quality data along the supply chain. Predictive models for future quality and shelf-life. Digital twin approach for quality simulation. Integration and validation across all tracks. |
| Track 4 Data Standards & Security
Projects: FreshAnalytics (GS1), FRED (semantics, ontologies, linked data), DataChainSec → EPCIS-based traceability. GS1 standards implementation. Semantic data models and ontologies for interoperability. Supply chain data security. |
The Research Timeline
Each project built on the last. The table below shows the progression from first research question to commercial product.
| Period | Project | What it contributed |
| 2018–2020 | FreshIndex | Established the core concept: dynamic shelf-life prediction for fresh meat based on cold chain data. Proved the commercial model. |
| 2019–2021 | Zukunftslabor | Developed the scientific shelf-life model for meat. Validated the underlying biology and modelling approach. |
| 2019–2022 | FreshAnalytics | Developed the NIR scanner. Built the data management platform. Field-validated at REWE Langel. Implemented GS1/EPCIS data standards. |
| 2020–2023 | FriDa | Extended quality data capture along the full supply chain for fresh fruit and vegetables. Partners: Euro Pool System, Uni Bonn, ATB Bremen, Fraunhofer IOSB. |
| 2021–2024 | FreshCloud | Cloud infrastructure for supply chain quality data. Scalable data management across multiple supply chain participants. |
| 2022–2025 | FreshTwin | Improved AI models using digital twin methodology. Predictive quality simulation across supply chain scenarios. |
| 2022–2025 | FRED | Integration of all tracks. Improved models. Semantic data standards, ontologies and linked data. Full system validation. |
| 2023–2026 | DataChainSec | Supply chain data security and integrity. EPCIS standards. tsenso as consortium participant. |
| 2025→ | newQC | Commercial product. Validated scanner. Guided inspection workflows. Consumer preference classifier. Origin pre-inspection. Audit-ready quality records. |
What the Research Produced: Validated Capabilities
The four research tracks produced capabilities that are now integrated into newQC. Each can be traced to a specific research programme with documented outcomes.
NIR Scanner — Non-Destructive Quality Measurement
Developed and calibrated across FreshAnalytics and FRED. Validated for seven fruit types. Field-deployed at REWE Langel in daily operational use. Models cover sweetness, acidity, juice content, firmness, ripeness, and remaining shelf-life.
Consumer Preference Classifier
Developed from FreshAnalytics scanner data. Validated through an in-store consumer study: 850 mango samples, 240 participants, 6 days, two locations in Germany. Classifier accuracy approximately 90%.
Supply Chain Quality Tracking
Developed across FriDa, FreshCloud, and FreshTwin. Tracks quality parameters from origin through the supply chain. Predicts future quality and shelf-life at defined thresholds. Digital twin modelling for scenario simulation.
Data Standards — EPCIS, GS1, Semantic Models
Implemented through FreshAnalytics (GS1 partnership) and FRED (ontologies, linked data, semantic interoperability). Supply chain data is structured to GS1 EPCIS standards, enabling integration with existing retailer and logistics systems.
Human Inspection Accuracy Baseline
Established through systematic in-store re-inspection of produce that had already passed commercial quality control: approximately 600 table grape units, 1,000 citrus units, and 1,500 apple units. Deviation rates in size conformity and defect assessment documented. Full methodology published in accompanying blog post.
