Shelf-Life Assessment of a Traditional Fried Snack: Moisture, Peroxide Value and Sensory Scores

What this study does. Karnataka's fried snacks — kodubale, chakli, thattai — sell everywhere from temple counters to online stores, yet their packets rarely carry a defensible 'best before' date. Shelf life is decided by two slow failures that run in parallel: moisture pickup, which turns crispness into chew, and oxidative rancidity of the frying oil absorbed into the snack, which turns the aroma stale long before anything looks wrong. This project measures both, plus what the customer actually perceives, over a real 60-day store.

One production batch of kodubale from a home-scale unit (fictional name) is split into two pack types — ordinary LDPE pouches and metallised laminate pouches — sealed the day of production and stored at ambient room conditions. At 0, 15, 30, 45 and 60 days, triplicate pouches of each type are opened and analysed: moisture by oven drying, peroxide value (PV) of the extracted fat by the FSSAI/AOAC iodometric titration, and free fatty acid acidity. In parallel, a 15-member semi-trained panel scores coded samples on the 9-point hedonic scale for appearance, aroma, crispness, taste and overall acceptability, with water and plain crackers between samples.

Shelf life is defined before the data exists: the first day on which any decisive parameter crosses its threshold — PV against the FSSAI limit for edible oils, sensory overall acceptability below the hedonic neutral point — for a given pack type. Comparing the two packages turns the project from a description into a question with a practical answer: does the better pouch buy the home-scale producer extra sellable weeks, and which measurement saw the failure first?

Home-scale producers of Karnataka's fried snacks sell products whose shelf life is folklore: a 'best before' on the label, if any, is copied from a neighbour's packet, and nothing connects it to measured moisture, oil quality or what customers actually taste. A producer cannot tell whether an ordinary LDPE pouch or a costlier metallised one changes the sellable weeks, and a returned stale packet teaches nothing because no baseline was ever measured.

At the student level, shelf-life projects often degrade into a single tasting at the end, or chemical numbers quoted without an end-point criterion, so the central shelf-life question — when does this product stop being sellable, and by which measure? — is never answered. Panels are assembled ad hoc, samples un-coded, criteria invented after seeing the data.

This project therefore runs a criteria-first, 60-day real-time assessment of one batch of kodubale in two pack types: chemical failure tracked by moisture and peroxide value against FSSAI limits, perceived failure tracked by a 15-member semi-trained hedonic panel, and shelf life defined as the first pre-declared threshold crossed — delivering the producer a defensible date and a packaging recommendation.

Abstract

This single-member study assesses the shelf life of kodubale, a traditional Karnataka fried snack, over 60 days of real-time ambient storage. One production batch from a home-scale unit is packed the same day in ordinary LDPE pouches and metallised laminate pouches. At 0, 15, 30, 45 and 60 days, triplicate pouches per pack type are analysed for moisture (oven drying), peroxide value and free fatty acid acidity of the extracted fat (FSSAI/AOAC iodometric and titration methods), while a 15-member semi-trained panel scores coded samples on the 9-point hedonic scale across five attributes. Pre-declared end-point criteria — the FSSAI peroxide-value limit and the hedonic neutral point for overall acceptability — define shelf life per pack type, and the two packages are compared for the weeks they add.

Introduction

Traditional fried snacks are a parallel food economy: home-scale and small units sell them daily with no laboratory behind them, and their 'best before' dates, where printed, are guesses. Kodubale is a rice-flour ring fried in oil, which makes it a clean model system — its two failure modes, moisture gain and oil oxidation, are exactly the ones food technology teaches, measurable with undergraduate equipment.

Existing knowledge and the gap

Rancidity chemistry and hedonic scaling are textbook material, and FSSAI publishes the analytical methods and the labelling law. What student projects usually omit is the design that makes a shelf-life claim: a single production batch (so batch variation does not masquerade as storage change), replicate pouches per time point, pre-declared end-point criteria, a panel with training and palate-cleansing discipline rather than whoever walks past, and a comparison against a relevant legal limit rather than 'it smelled odd'. The gap this project fills is a complete, criteria-first assessment that compares two realistic pack options for a home-scale producer and states which measurement — chemical or sensory — failed first.

Proposed study

1. Contract one production batch from the unit; record the recipe, frying oil and cooling; pack and seal 30 pouches per type the same day.
2. Fix the storage position, log ambient temperature and humidity through the study.
3. At each sampling day, open triplicate pouches per type; determine moisture, PV and FFA; photograph every sample.
4. Convene the 15-member panel per sampling day; score coded, randomised samples with palate cleansers; collect ballots.
5. Plot trends per pack type; apply the pre-declared criteria; identify the first failing parameter and day for each pack.
6. Report shelf life per pack type, the package comparison, and a plain-language recommendation for the producer.

Feasibility

• Technical: oven-dry moisture, iodometric PV titration and a hedonic panel are all B.Sc food-lab stock; the chemistry borrows the college's titration setup.
• Economic: snack, pouches and panel refreshments cost under ₹1,200 across the study; no instrumentation is bought.
• Operational: the panel's availability on five fixed afternoons is the logistical risk, solved by booking the calendar in advance. Ethics are simple but real: informed consent, ingredient/allergen disclosure, and no panelist with a relevant allergy. The main technical risks — PV endpoint subjectivity and moisture pickup during analysis — are handled with a blank-controlled titration and fast, closed weighing practice.

In scope

• One production batch, two pack types, five time points over 60 days, ambient storage only.
• Moisture, peroxide value, free fatty acid acidity, and a five-attribute hedonic panel of 15 semi-trained members.
• Criteria-first shelf-life determination per pack type and the package comparison.

Out of scope

• Accelerated storage (40 °C / Rancimat / Schaal oven) — a separate rig with its own validation.
• Microbial counts — a low-moisture fried snack's failures here are physicochemical and sensory; a micro-lab study would be a separate project.
• Water activity measurement and headspace oxygen analysis — listed under future scope.
• New recipe reformulation (antioxidants, frying-oil change) — the study measures, it does not redesign.

Research design: a real-time storage experiment with one batch, two packaging treatments, five destructive sampling points, chemical and sensory responses, and pre-declared end-point criteria. Individual study, 14 weeks at 8 h/week (the 60-day store sits inside weeks 4–13).

Phase 1 — Batch, packing and baseline (Weeks 1–3). One kodubale batch from the home-scale unit (fictional name); recipe, frying oil, fry temperature and cooling time recorded. Thirty pouches per type (LDPE; metallised laminate) filled to equal mass and heat-sealed the same day; 20 pouches per type held for the five time points plus spares. Day-0 analysis and panel run immediately: moisture, PV, FFA, and the sensory baseline.

Phase 2 — Storage (Weeks 4–13). Pouches stored in a single cupboard at ambient conditions; temperature and humidity logged daily with a cheap thermo-hygrometer. Pouches are not reopened between time points — sampling is destructive, three pouches per type per day.

Phase 3 — Sampling days (Days 15, 30, 45, 60). Each sampling day: triplicate pouches per type opened; moisture by oven drying to constant mass (105 °C); fat extracted from ground sample (solvent per lab practice) and PV by the FSSAI/AOAC iodometric method with a blank; FFA as % oleic acid by titration; samples photographed. Same afternoon: the 15-member semi-trained panel scores five attributes on the 9-point hedonic scale — coded three-digit samples, randomised pack order, water and plain crackers between samples, individual booths or spaced seating.

Panel discipline (Week 3): one orientation session on the scale and attributes, a practice round on fresh versus deliberately stale samples, consent forms with ingredient/allergen disclosure, and attendance booked for all five days.

Analysis plan (Weeks 13–14): mean ± SD per pack type per day for each chemical parameter; sensory medians per attribute; Friedman test across time points for overall acceptability per pack type; trend plots; application of the pre-declared criteria — PV versus the FSSAI edible-oil limit and overall acceptability below the hedonic neutral point — to identify the first failing parameter and its day; the pack comparison summarised as weeks gained.

Timeline: W1–3 batch, packing, baseline, panel orientation; W4–13 storage with sampling days; W13–14 analysis; W15-equivalent report, slides, viva within the 14-week envelope.

The study is a batch-to-verdict pipeline in which the shelf-life claim is generated by pre-declared criteria acting on measured trends — never by eyeballing the last sample. The workflow below is the report's methods spine.

flowchart TD
  A["One production batch: recipe, oil, process recorded"] --> B["Pack same day: 30 LDPE + 30 metallised pouches"]
  B --> C["Day-0 baseline: moisture, PV, FFA + sensory panel"]
  C --> D["Ambient storage; daily temp and humidity log"]
  D --> E{"Sampling day: 0, 15, 30, 45, 60"}
  E --> F["Open triplicate pouches per pack type"]
  F --> G["Moisture (oven), PV (iodometric), FFA (titration)"]
  F --> H["15-member hedonic panel, coded randomised samples"]
  G --> I{"Pre-declared criteria: PV <= FSSAI limit? overall >= neutral?"}
  H --> I
  I -->|Pass| J["Continue storage to next point"]
  I -->|Fail| K["Record first failing parameter and day"]
  K --> L["Shelf life per pack type; weeks gained by better pack"]
  L --> M["Producer recommendation and report"]

Data organisation

One row per pouch: pack type, sampling day, pouch code, mass, moisture %, PV (meq O₂/kg), FFA (% oleic acid), blank titre, analyst note; one ballot sheet set per panel day. Trends, criteria checks and the Friedman test regenerate from these tables; one PV calculation is hand-shown in the appendix with the blank correction. Sensory ballots are anonymised with member codes.

Core calculations (abridged)

Moisture % = (wet mass - dry mass) / wet mass x 100
PV (meq O2/kg fat) = (S - B) x N x 1000 / sample mass (g)
  S, B: sample and blank titres (mL); N: thiosulfate normality
End point 1: PV crosses the FSSAI edible-oil limit
End point 2: sensory overall acceptability < 5 (hedonic neutral)
Shelf life  = earliest day any pre-declared criterion fails

Results tables to fill (no numbers are pre-filled)

• Accelerated storage at elevated temperature or Rancimat induction time to correlate with the real-time result and shorten future studies.
• Water activity and headspace oxygen measurement per pouch to separate moisture failure from oxidation failure cleanly.
• Antioxidant trial: a natural-antioxidant (e.g., spice-extract) variant of the same snack, judged with the same criteria set.
• Frying-oil comparison: the same batch fried in two oils, tracking PV divergence — a natural sequel project.
• Micro-survey of label practices among local snack producers, giving the study a market-context chapter.
• A one-page producer card: the measured shelf life, the pack recommendation and storage advice, printable for the counter.