Milk Cooperative Fat-Based Payment Dispute
At the Kausalyaganga dairy cooperative in Khordha district, 41 members pour milk twice a day. The rate card reads: base price ₹7.00 per litre at 6.5% fat; for every 0.1 percentage point of fat above 6.5%, add ₹0.20 per litre; for every 0.1 point below, deduct ₹0.20. Yesterday's register shows Sasmita Pradhan 60 litres at 6.9%, Ramesh Behera 30 litres at 6.2%, and Bharati Das 55 litres at 6.5%; no other member poured. At the annual meeting the secretary says the society's average fat was “about 6.53%, so we are paying roughly the base rate”. Ramesh objects: his sample, he says, was 6.6%, not 6.2%, and he wants a re-test. The analyser's fee is ₹40 per sample, and a re-test means holding back payment for one day for all 41 members, several of whom have complained about delays before. The society's analyser log shows no calibration entry for the last four months.
1. Work out what the cooperative must pay each of the three members for the day, and the fat percentage it is actually paying on. Then explain what is wrong with the secretary's “6.53%”.
Read the worked answer
Rate per litre first, then multiply by litres.
- Sasmita: 6.9 − 6.5 = +0.4 points = 4 steps × ₹0.20 = +₹0.80 → ₹7.80/L × 60 L = ₹468.00
- Ramesh: 6.2 − 6.5 = −0.3 points = 3 steps × ₹0.20 = −₹0.60 → ₹6.40/L × 30 L = ₹192.00
- Bharati: exactly 6.5 → ₹7.00/L × 55 L = ₹385.00
Total litres = 145; total payout = ₹1,045.00.
The fat actually paid for is the litre-weighted average: (60×6.9 + 30×6.2 + 55×6.5) ÷ 145 = (414 + 186 + 357.5) ÷ 145 = 957.5 ÷ 145 = 6.60%.
The secretary's 6.53% is the plain mean of the three percentages, (6.9 + 6.2 + 6.5) ÷ 3. That treats Sasmita's 60 litres and Ramesh's 30 litres as equally important. Payment follows litres, so only the weighted figure describes the pool. A second error hides in the same sentence: ₹7.00 is not an “average” price at all, it is the price at exactly 6.5% fat. The average price actually paid is 1,045 ÷ 145 ≈ ₹7.21 per litre. Whenever an average is quoted, the weights must be stated.
2. Quantify Ramesh's claim. Should the secretary order a re-test? Set out the decision and the conditions attached to it.
Read the worked answer
Size of the claim. The difference is 6.6 − 6.2 = 0.4 points = 4 steps × ₹0.20 = ₹0.80 per litre. At 30 litres a day that is ₹24 a day, about ₹720 in a 30-day month, roughly ₹8,760 a year. The re-test costs ₹40 - under 2% of one month's disputed amount.
So the money alone does not decide it; the process does. An unchecked measurement dispute is read by 41 members as evidence that the analyser can be ignored, and every future reading becomes arguable.
Decision: order the re-test, but not on the disputed sample. A held sample can change on standing, and nobody can now prove the chain of custody. Instead:
- Draw a fresh duplicate at the pour point, sealed and counter-signed by Ramesh and the tester.
- Send the split sample to a second laboratory; keep one part in the society's custody.
- Record the analyser's calibration history - four months without a calibration entry is itself a finding.
Do not freeze all 41 payments. Pay Ramesh at the recorded 6.2% for the day and post the difference in a “disputed” column, settled after the result. Withholding everyone's money over ₹24 is a disproportionate response that punishes 40 uninvolved members.
3. Name the single procedural change that would stop this dispute from recurring, and say what makes it more effective than the alternatives.
Read the worked answer
A sealed duplicate sample drawn at the pour point, counter-signed by the member, with the analyser issuing a printed slip.
Why this one and not the others:
- It fixes the point of failure. The dispute is not about arithmetic; it is about whether the 6.2% reading came from Ramesh's milk at all. A counter-signed duplicate answers that question without argument.
- It is cheap: one extra bottle and a label. Compare with buying a second analyser, or with paying an analyst to sit through every pour.
- It creates an audit trail that works in both directions - the member is protected from a bad reading, and the society is protected from a claim invented after the milk has been pooled.
Pair it with a dated calibration entry logged weekly and displayed on the notice board. A duplicate sample on an uncalibrated machine only documents the error more precisely.