One workspace, six connected modules.
Each module works on the same products, datasets, and specifications — so analysis, formulation, quality, and modeling stay in the same analytical context.
- 01
Import
Upload a CSV of bench measurements; rows are validated before saving.
- 02
Analyse
Check every batch against your specification limits and flag deviations.
- 03
Formulate
Balance ingredients to 100% with USDA nutrient references and version history.
- 04
Assess
Predict shelf life from temperature, humidity and packaging with the live simulation engine.
- 05
Taste
Upload panel results; triangle and rating statistics are recalculated.
- 06
Report
Export a PDF or CSV built from the stored record, with reviewer decisions.
Every step works on the same products and batches, and each decision is recorded in the audit history. Suggestions are decision support — a qualified scientist approves or rejects them.
Food Analysis
Analyze nutritional, physicochemical, ingredient, and laboratory datasets against configurable specifications.
- Proximate and physicochemical parameters
- Observed vs reference comparison
- Deviation and status flags
- Analytical interpretation for review
Food Analysis — Whole Grain Bread, Batch 024
Formulation Intelligence
Compare formulations, model ingredient changes, and explore optimization scenarios side by side.
- Ingredient-level contribution
- Current vs baseline vs scenario
- Cost and nutrition trade-offs
- Functional role annotations
Formulation — High-Protein Snack Bar
Quality Intelligence
Analyze batch and QC data to identify trends, anomalies, and potential quality deviations.
- Batch and parameter tracking
- Specification ranges per product
- Trend and variation charts
- Anomaly review queue
Quality Intelligence — moisture by batch
Moisture values in the latest two batches are trending above the recent baseline. Outside configured specification — requires investigation.
Decision-support output — review before acting
Shelf-Life & Stability Simulation
A working simulation that estimates days until a product reaches critical water activity, from storage temperature, humidity and packaging barrier. Parameters are literature-typical starting values; confirm with real storage trials.
- Moisture uptake: a_w(t) = RH − (RH − a_w0)·e^(−k·t)
- Arrhenius temperature dependence (25 °C reference)
- Uncertainty range from activation energy ±10%
- Live in the platform today · method SL-1.0
Shelf-life simulator
Method SL-1.0 · runs live in your browser
Model: a_w(t) = RH − (RH − a_w0)·e^(−k·t), with k following the Arrhenius equation (25 °C reference). Rate constants and activation energies are illustrative literature-typical assumptions, not fitted to your data. Use as decision support; confirm with real storage trials before setting a shelf-life date.
Sensory Intelligence
Import panel results and analyse them with standard methods: hedonic ratings and ISO 4120 triangle tests.
- Mean, SD and 95% intervals per attribute
- Paired t-tests with Bonferroni adjustment
- Exact binomial triangle tests
- Stored, audited panel records
Sensory Intelligence — attribute profile
Panel mean, 0–9 scale · reference standard
R&D Workspace
Keep formulations, experiments, datasets, notes, analysis, and decisions organized in one environment.
- Experiment records and stages
- Dataset lineage
- Shared workspace roles
- Exportable reports
R&D Workspace — datasets
Evaluate each module on standard reference datasets.
The demo environment is clearly labelled and contains sample products, batches, and datasets.