A fraction of an industrial machine makes zero napkins.

Your analysis is spot on, the problem can be solved more efficiently with a large scale production. The capital expenditure cost however is similar (43M vs 50M). You point out they huge disparity in labor costs $25M vs .09M but labor is 'free' in India (which is to say there is so much of it available that the price is a lot less than elsewhere).

The incremental solution here wins because a small increment in cash instantly starts feeding the supply of napkins. The 'big machine' solution loses because it takes a huge investment to get to the point where you can make the napkins and nobody is willing to fund that.

There is another problem which are transportation costs. The infrastructure in India is, by all reports, spotty in the rural areas. By creating the manufacturing in the towns themselves you mitigate the transportation problems. That would not be possible with a centralized industrial machine. Further the folks who 'make' the napkins can also sell them so you have a personal relationship with someone in the town.

The incremental nature, the fact that India is labor rich and infrastructure poor, and the comparable capital costs make this an extraordinarily good solution to the problem.

Well, that assumes that this is the only individual in India that could be interested in the market segment. There are plenty of people in India for whom USD $500K is not a lot of money, particularly if a solid case can be made for the investment.

He could have invested four years trying to raise money to buy a number of industrial grade machines rather than doing what he did.

Here's another interesting data point. According to this site (http://www.euronuclear.org/info/encyclopedia/n/nuclear-power...) the average nuclear power plant in India produces 200MW of power. Deploying enough machines to provide napkins to the TAM would consume around 6% of the power output of a single nuclear power plant. An operation based around industrial machines would require about 4%. It's interesting when you can measure things in terms of the percentage utilization of a nuclear power plant.

I looked through the various responses to my post. Lots of good points. Of course, the infrastructure in India is something I cannot have a good grasp of from my vantage point in suburban Los Angeles. When faced with something like this I always fall back to one of my favorite Mark Twain sayings: "A man holding a cat by the tail learns something he can learn in no other way". This inventor held that cat by the tail. Hard to judge from the outside. I'll have to leave it at that.

And that person would be welcome to undercut him and provide better value. Before he came along, pads were too expensive. Now they're affordable. They also stimulate the local economy, both in terms of increased jobs and increased healthcare (very underrated). Regardless of the means, the end result is better.

And even if he did the industrial machine thing, you've missed the comment on transport issues - transportation is a big thing.

This problem is not new and I am sure that companies like J&J have looked at the Indian market. Reaching remote villages while providing a low price is something that a big company concerned about profits will not consider at all. Not until the numbers make sense.

His numbers for labor cost, namely $1/hour, are in the ballpark of free. It's probably a bit high (I'd guess something closer to 15-35rs/hour), but not unreasonable.

The minimum wage in India is 115rs/day, or 11.5rs/hour with a 10 hour workday, so it wouldn't even be legal to pay 5rs/hour (an order of magnitude less than the 50rs/hour algoshift postulated).

Assuming the machine operators earn a median Indian wage, assuming 8 hour days, 6 days/week, 52 weeks/year, their wage will be 28rs/hour.

$1 a day may seem extreme but in reality it's not quite as ridiculous as you suggest. In 2005, the World Bank estimated that 42% of India's population lived on $1.25 or less per day. In the interim years, the urban poor saw a 3.6% increase in income inequality and the rural 1.3%.

The issue of minimum wage in India is complicated, not all wage-earners get the either the national or state-level minimum wage (and women are disproportionally excluded). This article covers it in detail:

http://www.ilo.org/wcmsp5/groups/public/---ed_protect/---pro...

As you say, though, it still doesn't sound legal.

Procter and Gamble can already produce these things for a fraction of the cost. But building and maintaining a distribution network, marketing the product and making a healthy profit drives costs up.

This solution turns that problem on its head. Disperse the manufacturing locally, and let traditional/local social networks get the napkins distributed. In a country as large and complex as India, this is probably easier than finding 5,000 marketing campaigns to try to reach many small pockets of demand regarding a taboo subject.

But we wouldn't have it any other way.

As the article pointed out, the biggest problem in this case is not technical. It's social.

Now that you've shown that the big machines are more efficient overall, how do you get there from here? I would even say that if you know from the start that you want a bunch of $500K machines, then bootstrapping -- what he's doing now -- is in fact the easiest way. Google has some massive datacenters but they didn't start by spending $50M on hardware on day 1.

Wait a minute here, you multiplied the electrical cost per cheap machine by 175 since it makes 175 times less, then compared the cost of the 175 times more, then claimed it made 175 times less, but you already adjusted it to be 1 to 1. Are you double compensating here?

746W per machine => 7.46KWh per day. Multiply by $0.03 USD/KWh (=1.5 INR) to get $0.22/day or $0.00022 per napkin, slightly cheaper than the industrial solution. In other words, the household machines are about half as efficient with power but their cheaper rates on power more than compensates.

In terms of overall efficiency, other factors like raw materials, transportation, etc. are likely to dominate--the power expenditure is small for both options.

Unskilled/semiskilled laborers in India don't make anything close to that.

You are also missing the part where he is distributing production to poorer manufacturers who are willing to take a smaller profit, and thus pushing the market price down. Any commercial entity that buys the $500k machine will not want to get into a race to the bottom in pricing unless they are forced to do so. These distributed units which are run by poor women who don't want/need corporation level profits will force that race to the bottom. In the end, the big machines will likely prevail but prices will be much more reasonable.

>assuming five pads used per period

Assume 1-5 per DAY, at 3-6 days per 28 days. So, 3-30 per month, 40-400 per year per adolescent -> menopausal female.

Oh, and don't forget these are also useful for men in their old age.

I almost cried with pride reading this article. It is a massive, massive problem in the undeveloped world stopping girls and women from working and getting an education. By putting the means of production in the hands of a village collective, the results could be staggering.

See S.H.E. (Sustainable Health Enterprises) for another market-based example in this effort, based in Africa using plant matter without all this processing: http://www.sheinnovates.com/

In addition, it's important to note that this is merely a first revision. It's safe to assume, I think, that future revisions would improve upon the design and, due to the agility of manufacturing (how many industrial machines are manufactured per year?), it should be easier to prototype models that improve efficiency, cost and quality. I'd even go as far as to say that if this catches on it will eventually come close to matching the efficiency of an industrial machine for a fraction of the current cost.

Also to take in consideration is

- warehouse area to store raw materials - some kind of financial system to manage purchase orders / sales orders

21,000 napkins / hour. Let's assume there's 10 in a box, 8 boxes to a case and 30 cases in a pallet.

So that's 2400 napkins / pallet = ~ 9 pallets per hour. You're also going to need labelling, conveyors, people to pack the boxes, people to pack the cases, people to stack the pallets and then a few fork lifts to move the pallets plus of course engineering staff to maintain all this.

So it's not really small machine v big machine, it's small machine v production line which adds a lot of complexity.

The hint is the locality of production.

One end is a big centralized machinery designed to scale, and the appropriate logistics to ramp up availability across the country. But failures in production or transportation of goods will result in reduced availability. Also, centralized production drains the money to one place, namely the industrial area, and the local people might not have the money to continuously support buying these rags.

The other end is a distributed set of decentralized small-scale production fragments that produces local goods locally. The scheme will yield a higher local efficiency even if the total country-wide production of all units combine is more inefficient. It will recirculate local money more efficiently and boost local economy, again a local maxima, and this will allow more local people to buy the goods which was the whole point of this grassroots manufacturing. Also, production outages in one unit will not cause reduced availability since only a fraction of total production will be offline.

It gets fun to find analogies that fit this pattern. It's like Perforce vs Git, or West vs East.

Good!

The production is decentralized and it enables small business growth in rural areas.

The advantage here is they will be able to sell napkins to a community at a MUCH lower price and still make a profit using these small machines than the industrial ones.