AaharAnk v2.1 — Technical Specification and Validation Protocol

आहारांक

AaharAnk v2.1

Technical Specification and Validation Protocol

Open-Source Candidate Framework for Front-of-Pack Nutrition Labelling in India

Version 2.1 | August 2026 | Licence: CC BY 4.0

OPEN SOURCE — CC BY 4.0

Mandatory Disclaimer: AaharAnk v2.1 is a candidate technical specification. Grade thresholds are provisional and require calibration against a representative Indian product dataset of at least 500 products before use in production grading. The AaharAnk grade compares the nutritional composition of eligible packaged foods under the stated methodology. It is not medical advice, does not determine individual dietary suitability, and does not replace the full nutrition and ingredient declaration on the product label. AaharAnk is not a government-approved, validated, or mandated standard.

1. Definitions

Term	Definition
AaharAnk Score	The numerical result of subtracting total Positive Points (P) from total Negative Points (N). Lower score = more favourable nutritional composition profile.
AaharAnk Grade	One of five categories (अ, आ, इ, ई, उ) assigned to a product by mapping its AaharAnk Score to provisional grade threshold ranges.
Reference Amount	The basis on which nutritional values are calculated. Default: 100g for solid foods; 100ml for liquid foods. Category 6 uses reference portions (see Section 4.6).
Negative Points (N)	Points assigned based on nutrients to be limited: energy, sugars, saturated fat, and sodium.
Positive Points (P)	Points assigned based on nutrients and components to be encouraged: dietary fibre, protein, and fruits/vegetables/legumes percentage.
FVL%	Estimated percentage by weight of fruit, vegetables, legumes, nuts, and seeds in the product.
Protein Cap Rule	Inherited from FSA NPM 2004/05: if N ≥ 11 and FVL points < 5, protein points are excluded from P calculation. Exceptions apply (see Section 5.4).
Provisional Threshold	A grade boundary value proposed as a starting point for validation. Not suitable for production use until calibrated against an Indian product dataset.
NABL	National Accreditation Board for Testing and Calibration Laboratories (India). Accredited labs required for all testing.
NOVA-4	Ultra-processed food as classified by the NOVA system. Not used in the primary score; displayed as a secondary indicator.
NSS	Non-sugar sweetener. As defined by WHO 2023 guideline: all synthetic and naturally occurring or modified non-nutritive sweeteners not classified as sugars.¹
FSA NPM 2004/05	UK Food Standards Agency Nutrient Profiling Model 2004/05 (Rayner, Scarborough, Stockley). The declared baseline model for AaharAnk.²
CC BY 4.0	Creative Commons Attribution 4.0 International licence. Full terms: https://creativecommons.org/licenses/by/4.0/

2. Declared Baseline Model

AaharAnk v2.0 is explicitly derived from the UK Food Standards Agency Nutrient Profiling Model 2004/05, hereafter "FSA NPM 2004/05."

Citation: Rayner M, Scarborough P, Stockley L. Nutrient Profiles: Options for definitions for use in relation to food promotion and children's diets. London: Food Standards Agency, 2004.
URL: https://www.ndph.ox.ac.uk/food-ncd/files/about/uk-ofcom-nutrient-profile-model.pdf
Status: Published academic work in the public domain. Not patented or proprietary.
Validation: Independently reviewed by the UK Food Standards Agency (2009) and found "generally scientifically robust and fit for purpose."³ Used as the basis for NutriScore (EU) and UK Ofcom HFSS restrictions. Note: Health Star Rating is based on the Australian/NZ Nutrient Profiling Scoring Criterion, which was influenced by but differs from the FSA NPM 2004/05. The FSA review found the model fit for its specified UK application (food promotion to children) — this finding does not validate AaharAnk or its application in India.

AaharAnk does not claim to be an improvement over the FSA NPM 2004/05. It applies the same core model with declared, documented, and unvalidated modifications for Indian food categories. All departures are listed in Section 3 (Model Change Register).

Note on 2022/26 revised UK model: The UK published the outcome of its 2018 NPM review in January 2026, primarily updating the treatment of free sugars and fibre. AaharAnk v2.1 retains the 2004/05 model baseline, as it is the version most widely cited in peer-reviewed FOPL literature and forms the documented basis of NutriScore. A future version will evaluate alignment with the revised model pending FSSAI guidance on free sugars vs total sugars for Indian labelling.

3. Model Change Register

Every departure from the FSA NPM 2004/05 is declared below. The validation status column uses three values: Aligned with FSA NPM, Partially justified — requires validation, or Unvalidated — requires study.

#	Dimension	FSA NPM 2004/05	AaharAnk v2.0	Reason	Status
1	Protein maximum	0–5 pts (FSA NPM 2004/05 official)	0–5 pts	Retained as-is. Note: some implementations show 7 pts — the official FSA NPM 2004/05 document caps protein at 5 points.	✓ Aligned
2	Sugars component	Total sugars per 100g	Total sugars per 100g	Retained. Note: 2022 UK review moved to free sugars. AaharAnk defers this pending FSSAI guidance.	✓ Aligned (2004/05)
3	Saturated fat (Cat 3)	Absolute g/100g	Ratio: (Sat Fat / Total Fat) × 100	Better differentiation between Indian oils with very different fatty acid profiles but similar total fat	✗ Unvalidated
4	Condiment evaluation (Cat 6)	Per 100g reference amount	Per reference portion (5–20g, food-type specific)	Indian spices/condiments consumed in 5–20g quantities. Per-100g creates misleading scores for high-concentration foods used in small amounts. Reference portions from ICMR/Codex.	⚠ Partially justified
5	Millet bonus	Not present	+2 P points (if eligible)	Millets are nutritionally superior to refined grains across multiple parameters per ICMR data. [Exact ICMR/NIN publication citation to be inserted at V-05 — see Source Verification Ledger.] Supports India's national millet promotion. Eligibility: ≥50% whole millet by weight, fibre score ≥2, not NOVA-4.	✗ Unvalidated
6	Fermented food bonus	Not present	+1 P point (restricted scope)	Live-culture fermented foods (plain dahi, traditional brine pickles) have been associated with gut microbiome benefits in observational studies, though evidence in Indian populations specifically is limited.^5a Restricted to plain variants with live cultures declared on label. Does not apply to sweetened, flavoured, or vinegar-acidified variants.	✗ Unvalidated
7	Sweetener treatment	Not present	Removed from primary score. Moved to "Contains NSS" disclosure indicator.	WHO 2023 recommends against NSS for weight control.¹ Disclosure preferred over grade penalty to avoid discouraging reformulation away from added sugars.	⚠ Scientifically supported disclosure; penalty approach unvalidated
8	NOVA processing	Not present	Removed from primary score. Moved to "High Processing" secondary indicator.	NOVA inter-rater reliability issues and lack of an Indian-specific decision tree make primary score incorporation premature. Secondary display maintains transparency.	⚠ Partially justified
9	FVL qualifying ingredients	Fruit, vegetables and nuts (FSA 2004/05)	Fruit, vegetables, nuts, legumes, and seeds	Indian dietary pattern includes significant legume and seed consumption. Addition of legumes and seeds extends coverage to key Indian food groups.	✗ Unvalidated
10	Protein cap exemptions	Cap applies to all foods when N≥11 and FVL<5	Cap does not apply to Category 4 (Dairy) or Category 5 (Nuts)	Protein is the defining nutritional characteristic of these categories. Excluding it would produce misleading within-category comparisons.	✗ Unvalidated
11	Beverage grading and water override	FSA NPM binary pass/fail for marketing to children — not a five-grade system	Separate stricter five-grade thresholds for beverages; plain water auto-grades अ	Five-grade beverage system adapted from Nutri-Score and Singapore Nutri-Grade beverage approach.	✗ Unvalidated
12	Fibre thresholds	FSA NPM 2004/05 AOAC: approximately ≤0.9g=0, >0.9=1, >1.9=2, >2.8=3, >3.7=4, >4.7=5	AaharAnk: ≤3.0g=0, >3.0=1, >3.7=2, >4.4=3, >5.1=4, >5.8=5	Thresholds raised to reflect Indian food fibre profile — Indian staples (dal, atta, millets) have higher fibre content than processed European foods the FSA model was calibrated against.	✗ Unvalidated
13	Grade thresholds	NutriScore: A ≤−1, B 0–2, C 3–10, D 11–18, E ≥19 (general foods)	Provisional: अ ≤0, आ 1–3, इ 4–11, ई 12–20, उ ≥21	Provisionally adjusted (without empirical calibration) after correcting score range (max N=40, not 47 as stated in v1.0). Must be validated against a representative Indian product dataset of ≥500 products before use.	✗ Unvalidated

4. Food Classification System

Before calculating a score, every product must be assigned to exactly one of seven food categories. The category determines which algorithm variant applies. Classification is by primary ingredient by weight (>50% rule).

#	Category	Classification Rules	Special Algorithm
1	General Foods	Default for all solid foods not classified below	Standard algorithm
2	Beverages	Liquids intended for drinking. Exception: plain milk and plant-based drinks where milk/plant base >50% by weight → Category 4	Stricter thresholds; only plain water auto-grades अ
3	Fats, Oils & Ghee	Fat/oil is primary ingredient (>50%)	Sat fat ratio scoring replaces absolute sat fat
4	Dairy & Paneer	Milk or milk-derived product where milk is primary	Protein always counted (no protein cap applies)
5	Nuts, Seeds & Dry Fruits	Whole or processed nuts, seeds, dried fruits as primary ingredient	Sat fat ratio scoring; protein always counted
6	Masalas, Spices & Condiments	Spice, spice blend, condiment, pickle, chutney. Typical serving <25g.	Evaluated per reference portion, not per 100g
7	Millets & Traditional Grains	Whole or minimally processed millet or traditional grain ≥50% by weight	Millet bonus +2 P points (eligibility check required)

Multiple-ingredient products: If a product spans two categories, classify by the primary ingredient by weight. If the primary ingredient comprises >50% by weight, its category takes precedence. If no single ingredient exceeds 50%, classify as Category 1 (General Foods) and document the classification rationale.

4.6 Category 6 — Reference Portions (India-Specific)

All reference portions below are provisional research values. Exact documentary citations (specific publication, edition, table and page) for each value are outstanding and must be confirmed at V-05. Until verified, these should be treated as proposed starting points for the pilot study, not established regulatory reference amounts.

Product Type	Proposed Portion	Proposed Source (to be verified)
Ground single spices (haldi, chilli, coriander, cumin, pepper)	5g	ICMR dietary guidelines [exact document, year, page outstanding]
Spice blends (garam masala, sambhar, biryani, pav bhaji masala)	10g	Recipe-based estimate for 4-serving Indian preparation [not an authoritative source — requires ICMR/Codex validation]
Pickles, achaar (all varieties)	15g	National Nutrition Monitoring Bureau [exact survey, year, table outstanding]
Condiments (ketchup, soy sauce, vinegar)	15g	Codex Alimentarius [exact standard number and table outstanding]
Honey	20g	FSSAI labelling guidance [exact notification outstanding]
Chutneys (mint, tamarind, coconut)	20g	Recipe-based estimate [not an authoritative source — requires validation]

Category 6 scoring procedure: Obtain lab-tested nutritional values per 100g. Multiply each nutrient value by (reference portion ÷ 100) to convert to per-portion values. Apply the standard point-allocation tables directly to these per-portion values — do not convert back to per-100g. The scores are thus computed on a per-portion basis, reflecting actual dietary exposure rather than a per-100g analytical equivalent. This is an AaharAnk-specific adaptation and is marked as unvalidated in the Model Change Register.

5. Complete Algorithm

5.1 Core Formula

AaharAnk Score = N − P

Corrected score range: Theoretical minimum −18 (Category 7 product with all bonuses, P=18, N=0). Theoretical maximum +40 (maximum N=40, P=0). No operational range is stated — no Indian product dataset has been analysed to establish one.

5.2 Negative Components — Full Point Allocation Tables

Apply to values per 100g (solid foods) or per 100ml (liquid foods).

A. Energy (kJ per 100g/100ml)

0 pts	1 pt	2 pts	3 pts	4 pts	5 pts	6 pts	7 pts	8 pts	9 pts	10 pts
≤335	>335–≤670	>670–≤1005	>1005–≤1340	>1340–≤1675	>1675–≤2010	>2010–≤2345	>2345–≤2680	>2680–≤3015	>3015–≤3350	>3350

B. Total Sugars (g per 100g/100ml)

0 pts	1 pt	2 pts	3 pts	4 pts	5 pts	6 pts	7 pts	8 pts	9 pts	10 pts
≤4.5	>4.5–≤9.0	>9.0–≤13.5	>13.5–≤18.0	>18.0–≤22.5	>22.5–≤27.0	>27.0–≤31.0	>31.0–≤36.0	>36.0–≤40.0	>40.0–≤45.0	>45.0

C. Saturated Fat (g per 100g/100ml) — Categories 1, 2, 4 only

0 pts	1 pt	2 pts	3 pts	4 pts	5 pts	6 pts	7 pts	8 pts	9 pts	10 pts
≤1.0	>1.0–≤2.0	>2.0–≤3.0	>3.0–≤4.0	>4.0–≤5.0	>5.0–≤6.0	>6.0–≤7.0	>7.0–≤8.0	>8.0–≤9.0	>9.0–≤10.0	>10.0

C2. Saturated Fat Ratio (% of total fat) — Categories 3 and 5 only

Calculate: SFR = (Saturated Fat ÷ Total Fat) × 100

0 pts	2 pts	4 pts	6 pts	8 pts	10 pts
≤10%	>10%–≤20%	>20%–≤30%	>30%–≤40%	>40%–≤60%	>60%

Unvalidated: SFR thresholds are provisional. Intermediate values: assign points from the band the value falls into — no interpolation. Fractions are rounded to one decimal place before band assignment. This rule applies identically to all point tables in this document.

D. Sodium (mg per 100g/100ml)

0 pts	1 pt	2 pts	3 pts	4 pts	5 pts	6 pts	7 pts	8 pts	9 pts	10 pts
≤90	>90–≤180	>180–≤270	>270–≤360	>360–≤450	>450–≤540	>540–≤630	>630–≤720	>720–≤810	>810–≤900	>900

Maximum N = Energy(10) + Sugars(10) + SatFat/SFR(10) + Sodium(10) = 40 pts

5.3 Positive Components — Full Point Allocation Tables

E. Dietary Fibre (g per 100g/100ml)

Declared unvalidated modification: The fibre thresholds below (AOAC method, 3.0–5.8g scale) differ from the FSA NPM 2004/05 AOAC thresholds as published (approximately 0.9–4.7g scale). AaharAnk uses higher thresholds reflecting the range of fibre content in Indian foods. This deviation is declared as an unvalidated modification — see Model Change Register row 13. The thresholds will be reviewed during dataset calibration (V-03).

0 pts	1 pt	2 pts	3 pts	4 pts	5 pts
≤3.0	>3.0–≤3.7	>3.7–≤4.4	>4.4–≤5.1	>5.1–≤5.8	>5.8

AaharAnk adaptation: thresholds set at higher values to reflect Indian food fibre profile. Provisional — requires validation during V-03.

F. Protein (g per 100g/100ml)

0 pts	1 pt	2 pts	3 pts	4 pts	5 pts
≤1.6	>1.6 to ≤3.2	>3.2 to ≤4.8	>4.8 to ≤6.4	>6.4 to ≤8.0	>8.0

Maximum protein = 5 pts (FSA NPM 2004/05). Subject to protein cap rule (Section 5.4).

G. Fruits, Vegetables, Legumes, Nuts & Seeds (FVL) — % by weight

Declared modification from FSA NPM 2004/05: The original FSA model awards FVL points at only four levels: 0 (≤40%), 1 (>40%), 2 (>60%), 5 (>80%). AaharAnk v2.1 retains these four levels exactly. The intermediate 3-pt and 4-pt bands in v2.0 were an error and have been removed.

0 pts	1 pt	2 pts	5 pts
≤40%	>40% to ≤60%	>60% to ≤80%	>80%

Note: FVL% includes fruit, vegetables, nuts, and legumes as defined in FSA NPM 2004/05. Legumes and seeds are included in AaharAnk as a declared extension of the qualifying ingredients list (see Model Change Register row 10).

H. Millet Bonus (Category 7 only)

Add +2 P points if ALL of the following are true:

Product classified as Category 7 (Millets & Traditional Grains)
Whole millet content ≥50% by weight (declared on label or verified by ingredient list)
Fibre score (Table E) ≥ 2 points
Product is NOT classified NOVA-4

I. Fermented Food Bonus

Add +1 P point if ALL of the following are true:

Product is plain dahi/yogurt with live cultures declared on label at end of shelf life, OR traditional brine-fermented pickle (not vinegar/acetic acid-based)
No added sugar or sweetener
Not flavoured or modified with additional ingredients

Maximum P = Fibre(5) + Protein(5) + FVL(5) + Millet Bonus(2, if eligible) + Fermented(1, if eligible) = 18 pts maximum (Category 7 eligible product)

5.4 Protein Cap Rule (inherited from FSA NPM 2004/05)

Condition: If calculated N ≥ 11 AND FVL points < 5

Then: Protein points are excluded from P total

Formula becomes: P = Fibre + FVL (protein excluded)

Exceptions — protein always counted regardless:

Category 4 (Dairy & Paneer)
Category 5 (Nuts, Seeds & Dry Fruits)

Rationale: The protein cap prevents products high in sugar, fat, or sodium from achieving a more favourable grade simply by adding protein. Exceptions for dairy and nuts reflect that protein is a defining nutritional characteristic of these categories and excluding it would produce misleading comparisons within the category.

6. Secondary Indicators

Secondary indicators appear alongside the AaharAnk badge but are not incorporated into the grade score.

Indicator	Trigger	Display Text	Scientific Basis
NSS Disclosure	Any NSS present in ingredient list (acesulfame K, aspartame, sucralose, stevia/steviol glycosides, saccharin, neotame, advantame, cyclamates)	"Contains Non-Sugar Sweetener"	WHO's 2023 conditional recommendation advises against using NSS as a means of achieving weight control or reducing NCD risk.¹ The recommendation does not apply to individuals with pre-existing diabetes. The proposed disclosure format itself remains unvalidated as a FOPL tool.
High Processing	Classification under a validated NOVA decision protocol applied to the full ingredient list. Note: the "High Processing" indicator is not operational until a validated Indian-specific NOVA decision tree is published. The ingredient examples listed here (hydrogenated oils, modified starches, etc.) are illustrative of NOVA-4 markers, not a definitive classification rule. A single listed ingredient does not automatically trigger classification.	"High Processing"	NOVA processing level is associated with adverse dietary patterns. Displayed separately because NOVA classification inter-rater reliability requires a validated Indian decision tree before incorporation into the primary score.

7. Pseudocode

The following pseudocode is provided so that two independent assessors applying the same lab data should reach an identical AaharAnk Score. Any implementation must produce identical results for all worked examples in Section 8.

FUNCTION calculate_aaharank_v21(product):

  // STEP 1: Classify product into one of 7 categories
  category = classify_category(product)

  // STEP 2: Determine reference amount
  IF category == 6:
    ref_amount = get_reference_portion(product.type)
    // Convert to per-100g equivalent before scoring
    scale = ref_amount / 100
    energy = product.energy_kj * scale
    sugars = product.total_sugars_g * scale
    sat_fat = product.sat_fat_g * scale
    sodium = product.sodium_mg * scale
    fibre = product.fibre_g * scale
    protein = product.protein_g * scale
  ELSE:
    // Use per-100g or per-100ml values as supplied by NABL lab
    energy = product.energy_kj_per_100
    sugars = product.total_sugars_g_per_100
    sat_fat = product.sat_fat_g_per_100
    sodium = product.sodium_mg_per_100
    fibre = product.fibre_g_per_100
    protein = product.protein_g_per_100

  fvl_pct = product.fvl_percent

  // STEP 3: Calculate Negative Points
  N_energy = lookup_energy_pts(energy)        // Table A
  N_sugars = lookup_sugars_pts(sugars)        // Table B

  IF category IN [3, 5]:
    IF product.total_fat_g > 0:
      sfr = (sat_fat / product.total_fat_g) * 100
    ELSE:
      sfr = 0
    N_satfat = lookup_sfr_pts(sfr)            // Table C2
  ELSE:
    N_satfat = lookup_satfat_pts(sat_fat)     // Table C

  N_sodium = lookup_sodium_pts(sodium)        // Table D

  N = N_energy + N_sugars + N_satfat + N_sodium
  // No sweetener penalty in primary score
  // No NOVA penalty in primary score

  // STEP 4: Calculate Positive Points
  P_fibre = lookup_fibre_pts(fibre)           // Table E
  P_protein = lookup_protein_pts(protein)     // Table F
  P_fvl = lookup_fvl_pts(fvl_pct)            // Table G

  // STEP 5: Apply Protein Cap Rule (inherited from FSA NPM 2004/05)
  IF N >= 11 AND P_fvl < 5:
    IF category NOT IN [4, 5]:               // Cat 4, 5 exempt
      P_protein = 0                          // Exclude protein

  // STEP 6: Apply India-specific bonuses
  P_millet = 0
  IF category == 7:
    IF (product.whole_millet_pct >= 50 AND
        P_fibre >= 2 AND
        product.nova_group != 4):
      P_millet = 2

  P_fermented = 0
  IF (product.type IN ["plain_dahi", "brine_pickle"] AND
      product.live_cultures == TRUE AND
      product.added_sugar == FALSE AND
      product.flavoured == FALSE):
    P_fermented = 1

  P = P_fibre + P_protein + P_fvl + P_millet + P_fermented

  // STEP 7: Calculate final score
  score = N - P

  // STEP 8: Map score to grade (provisional thresholds)
  grade = lookup_grade(score, category)

  // STEP 9: Determine secondary indicators
  nss_flag = any_nss_in_ingredients(product.ingredients)
  nova_flag = (product.nova_group == 4)

  RETURN {
    score: score,
    N: N,
    P: P,
    grade: grade,
    nss_disclosure: nss_flag,
    high_processing: nova_flag
  }

8. Worked Calculation Examples

Important: All product examples below are illustrative only. The nutritional values used are representative estimates, not lab-tested figures. For the pilot phase, grades may be computed from verified manufacturer-declared nutrition data (as legally required under FSSAI Labelling Regulations 2020), supplemented by NABL-accredited laboratory testing for a representative audit sample. Full NABL testing for all products is recommended before any production-scale grading or mandatory labelling scheme. These examples demonstrate the calculation procedure, not production grades.

8.1 Illustrative Example: Whole Wheat Atta (1 kg)

Category 1 (General Foods). Values per 100g — illustrative only.

Energy: 1420 kJ → 4 pts

Sugars: 2.0g → 0 pts

Sat Fat: 0.5g → 0 pts

Sodium: 5mg → 0 pts

N = 4 + 0 + 0 + 0 = 4

Fibre: 9.0g → 5 pts | Protein: 12.0g → 5 pts (capped at FSA NPM max) | FVL%: 0% → 0 pts

Protein cap check: N(4) < 11 → protein counted. P = 5 + 5 + 0 = 10

Score = 4 − 10 = −6 → Provisional Grade: अ (Higher Quality) [≤0]

8.2 Illustrative Example: Potato Chips (150g)

Category 1 (General Foods) + NOVA-4 flag. Values per 100g — illustrative only.

Energy: 2230 kJ → 7 pts

Sugars: 1.2g → 0 pts

Sat Fat: 4.5g → 4 pts

Sodium: 480mg → 5 pts

N = 7 + 0 + 4 + 5 = 16

Fibre: 3.8g → 1 pt | Protein: 6.0g → 3 pts (pre-cap) | FVL%: 55% → 1 pt (>40% to ≤60% = 1 pt under corrected FSA NPM table)

Protein cap check: N(16) ≥ 11 AND FVL pts(1) < 5 → protein excluded

P = 1 + 0 + 1 = 2

Score = 16 − 2 = 14 → Provisional Grade: ई (Lower Nutritional Quality) [12–20]

Secondary indicators: High Processing flag (NOVA-4).

8.3 Illustrative Example: Garam Masala (100g) — Per-Portion Method

Category 6 (Masalas). Evaluated per 10g reference portion. Values per 100g from lab — illustrative only.

N = 0 + 0 + 0 + 0 = 0

Fibre: 1.49g → 0 pts | Protein: 1.26g → 0 pts | FVL%: 100% → 5 pts

Protein cap: N(0) < 11 → protein counted. P = 0 + 0 + 5 = 5

Score = 0 − 5 = −5 → Provisional Grade: अ (Higher Quality) [≤0]

Important limitation: Per-portion scoring for Category 6 products can produce near-zero negative scores for highly concentrated foods (e.g., high-sodium pickle pastes, seasoning mixes) that may contribute meaningfully to daily sodium intake across multiple servings. This is a known limitation of reference-portion scoring. The sodium contribution of condiments and spices should be assessed cumulatively. The Category 6 method remains unvalidated and subject to revision after V-05.

8.4 Illustrative Example: Pure Cow Ghee (500ml)

Category 3 (Fats, Oils & Ghee). Sat fat ratio scoring. Values per 100g — illustrative only.

Total Fat: 99.5g | Sat Fat: 61.9g | SFR = (61.9/99.5) × 100 = 62.2% → 10 pts

Energy: 3696 kJ → 10 pts | Sugars: 0g → 0 pts | Sodium: 2mg → 0 pts

N = 10 + 0 + 10 + 0 = 20

Fibre: 0g → 0 pts | Protein: 0.3g → 0 pts | FVL%: 0% → 0 pts

P = 0

Score = 20 − 0 = 20 → Provisional Grade: ई (Lower Nutritional Quality) [12–20]

Threshold boundary note: Score of 20 falls within ई (12–20) under provisional thresholds. This illustrates how the boundary between ई and उ will require careful calibration — ghee sits close to that boundary and small threshold adjustments could shift it. Until threshold validation is complete, this illustrative grade should be treated as approximate.

9. Provisional Grade Thresholds

These thresholds must not be used for production pack grading until validated against a minimum 500-product Indian dataset. They are published here for transparency and scientific review only.

9.1 General Foods (Category 1)

अ Higher Quality	आ Good Quality	इ Moderate	ई Lower Quality	उ Limit Frequency
≤ 0	1 to 3	4 to 11	12 to 20	≥ 21

9.2 Beverages (Category 2 — stricter)

अ	आ	इ	ई	उ
Water only (BIS-certified, no additives)	≤ 1	2 to 4	5 to 8	≥ 9

9.3 All Other Categories

Categories 3, 4, 5, 6, 7 use the same provisional thresholds as General Foods (Section 9.1), with their respective algorithm modifications applied first. Category-specific thresholds to be developed during the validation programme.

10. Validation Protocol

The following ten gates must be completed before AaharAnk v2.1 thresholds are confirmed and before any production-grade labelling commences.

Gate	Requirement	Standard	Owner
V-01	Mathematical proof: verify score range and threshold coverage with no gaps	Computer-verified calculation across all threshold boundary values	Technical Secretariat
V-02	Algorithm reproducibility: two independent assessors reach identical score for same lab data	Zero discrepancy across 20 test products	Scientific Advisory Council
V-03	Dataset calibration: compute scores for ≥500 representative Indian products across all 7 categories	Representative sample validated by food scientist. Dataset published openly.	Scientific Advisory Council
V-04	Threshold review: assess whether provisional thresholds correctly classify products across all categories	Peer review by minimum 2 independent food scientists	Scientific Advisory Council
V-05	Indian food category validation: confirm Category 6 reference portions and Category 7 millet eligibility criteria against real products	Comparison against ICMR and Codex reference data	Nutrient Profiling Working Group
V-06	Consumer comprehension study: ≥400 respondents across literacy levels, regions, and languages	Published study report. ≥70% correct identification of अ as most favourable, उ as least favourable	Consumer Testing Working Group
V-07	Legal IP review: confirm CC BY 4.0 application, trademark approach, and open-source model are legally sound under Indian IP law	Written opinion from qualified Indian IP counsel	Legal Counsel
V-08	FSSAI alignment check: confirm AaharAnk methodology is compatible with existing FSSAI Food Safety and Standards (Labelling and Display) Regulations 2020	Written assessment by food regulatory lawyer	Legal Counsel
V-09	Conflict-of-interest audit: all SAC members declare funding sources	Published declarations on public repository	COI Committee
V-10	Version lock: publish version 2.1 final specification with SHA-256 checksum	Immutable public record of the validated specification	Technical Secretariat

11. Open Governance

Methodology licence: Creative Commons Attribution 4.0 (CC BY 4.0). https://creativecommons.org/licenses/by/4.0/
Name and badge design: Trademark protection (pending filing) to ensure accurate use. No licensing fee for compliant use. Modified versions must use a different name and badge design.
Version control: Every published version carries a version number and SHA-256 checksum of the specification document. Users must declare which version they applied.
Public repository: Methodology, point tables, pseudocode, dataset, worked examples, and SAC minutes published at a public URL with version control. URL to be confirmed at v2.0 final release.
Update cycle: Annual review by Scientific Advisory Council. Minor updates (threshold adjustments): 30-day public comment, 90-day regrade period. Major updates (algorithm changes): 90-day public comment, 180-day transition.
Conflict of interest: Industry funding must be declared by all SAC members. Industry representatives may participate as observers, not voting members.

12. Source Verification Ledger

#	Claim	Source	URL	Verified
1	NSS definition includes stevia, aspartame, sucralose, saccharin, acesulfame-K	WHO 2023 NSS guideline	https://www.who.int/news/item/15-05-2023-who-advises-not-to-use-non-sugar-sweeteners-for-weight-control	✓
2	FSA NPM 2004/05 algorithm basis	Rayner et al. 2004, Oxford/FSA	https://www.ndph.ox.ac.uk/food-ncd/files/about/uk-ofcom-nutrient-profile-model.pdf	✓
3	FSA NPM found "generally scientifically robust" by independent panel	UK FSA Board Minutes, 25 March 2009	http://www.food.gov.uk/multimedia/pdfs/board/boardmins090325.pdf	✓
4	Millets have 2–3× calcium, iron, fibre vs refined grains	ICMR nutritional data; requires specific ICMR publication citation	[Specific ICMR publication to be cited in V-05]	⚠ Pending
5	NutriScore is trademarked by Santé Publique France	NutriScore official governance	https://www.santepubliquefrance.fr/determinants-de-sante/nutrition-et-activite-physique/nutri-score	✓
6	SC proceedings on FOPL PIL	lawbeat.in reporting	https://lawbeat.in/top-stories/supreme-court-gives-fssai-last-chance-on-front-of-pack-warning-labels-for-high-sugar-salt-fat-foods-1621897	✓ (news report; official order text pending)
7	ICMR-INDIAB diabetes/prediabetes estimates	Lancet Diabetes Endocrinol 2023;11:474–489	DOI: 10.1016/S2213-8587(23)00119-5	✓

13. Limitations and Unresolved Questions

Grade thresholds are provisional: The thresholds in Section 9 have not been calibrated against Indian product data. They may need significant adjustment after V-03.
Sat fat ratio scoring (Categories 3, 5): Not validated. The boundary values in Table C2 are proposed based on general knowledge of Indian oil fatty acid profiles, not a systematic study.
Millet and fermented food bonuses: Not validated. Both are based on general nutritional science reasoning. The specific bonus magnitudes (+2, +1) are arbitrary pending validation.
FVL% estimation: The calculation method for FVL% in composite and processed foods is not fully specified in this version. A supplementary FVL calculation guide will be published before V-05.
NOVA classification decision tree: No India-specific NOVA-4 decision tree exists. Inter-rater reliability of NOVA classification for Indian foods is unestablished. The "High Processing" secondary indicator should be interpreted cautiously until this is resolved.
Free sugars vs total sugars: AaharAnk v2.0 uses total sugars (aligned with FSA NPM 2004/05). The 2022 UK review moved to free sugars. FSSAI guidance on which sugar measurement to use for Indian FOPL will determine whether AaharAnk needs to align in a future version.
Category-specific thresholds: Only general food thresholds are specified in v2.0. Category-specific thresholds for beverages, fats/oils, dairy, and other categories require dedicated validation studies.
Devanagari literacy: Consumer comprehension across literacy levels, regions, and non-Hindi-speaking populations is unestablished. V-06 (comprehension study) is mandatory before making any claim about pan-Indian applicability.

14. Version Control and Change Log

Version	Date	Changes from Previous
v1.0	April 2026	Initial draft. Sweetener penalty (+4) in primary score. NOVA penalty (+3) in primary score. Maximum N = 47. Thresholds inherited directly from NutriScore. Open-source declaration without CC BY 4.0 specification. No formal model change register.
v2.1	August 2026	Sweetener penalty moved to secondary indicator (NSS disclosure). NOVA penalty moved to secondary indicator (High Processing). Maximum N corrected to 40. Grade thresholds provisionally adjusted without empirical calibration for corrected score range and marked provisional. Formal Model Change Register added. Full point-allocation tables included. Pseudocode added. Validation protocol (10 gates) added. Source verification ledger added. Limitations section added. Open-source model specified as CC BY 4.0. Health claim language replaced with neutral nutritional composition language. Supreme Court framing corrected to reflect news reporting rather than official order text.

References

¹ World Health Organization. Use of non-sugar sweeteners: WHO guideline. Geneva: WHO; 2023. https://www.who.int/news/item/15-05-2023-who-advises-not-to-use-non-sugar-sweeteners-for-weight-control

² Rayner M, Scarborough P, Stockley L. Nutrient Profiles: Options for definitions for use in relation to food promotion and children's diets. London: Food Standards Agency, 2004. https://www.ndph.ox.ac.uk/food-ncd/files/about/uk-ofcom-nutrient-profile-model.pdf

³ UK Food Standards Agency Board Minutes, 25 March 2009. http://www.food.gov.uk/multimedia/pdfs/board/boardmins090325.pdf

^5a Marco ML et al. Health benefits of fermented foods: microbiota and beyond. Current Opinion in Biotechnology. 2017;44:94–102. DOI: 10.1016/j.copbio.2016.11.010. Note: this citation supports the general association; Indian-population-specific evidence for this bonus requires documentation at V-05.

⁴ ICMR/National Institute of Nutrition. Nutritional data on Indian millets. [Specific publication: likely "Nutritive Value of Indian Foods," NIN/ICMR — exact edition, year and page to be confirmed at V-05 validation gate before the numerical 2–3× claim is reinstated.]