RSS
 

Posts Tagged ‘quality control’

Quality Control

20 Nov

When I decided that I needed a distraction-free writing studio, I built one in my basement. I brought together a set of shelves that I had and configured them into a square. By hanging fabric around the inside of the square and putting a comfortable chair, a footstool, and an upside down wastebasket to rest my tea mug on inside the square, I made myself quite a comfortable 4′x4′ studio with no visual or auditory distractions. One side effect of this setup is as that as I’m writing this, I’m surrounded by a large percentage of my personal belongings. From the inside I just see fabric, but I still know that I am surrounded by craft materials, old papers, and reasonably useless appliances.

It’s about the appliances that I want to talk now. Just off the top of my head, I know that there’s an ice shaver, a couple of crock-pots, a toaster, and a space heater. None of them work. The crock-pots, toaster, and space heater have all stopped working, and the ice shaver never worked in the first place. I can’t bring myself to throw them away, because they all seem like they might have useful parts in them, so in the mean time they just hang out on shelves in my basement. I’ve only been living here for three years, and I’m a fairly frugal person, and yet I’ve already managed to build up a collection of five dead appliances.

I think this says something important, and a bit disturbing, about our modern culture. Upstairs in the kitchen, my mother has a blender that she bought shortly after I was born. It still works, and I use it to make milkshakes for myself and my grandfather several times a week. In the two years that I lived in Chicago, I went through three blenders just making milkshakes for myself.

It’s a cliche, but they don’t make anything like they used to. I don’t think I paid more than $20 for any of those blenders in Chicago. But in total, at 3x$20 every 2 years, I was paying $30 a year in blender purchases. Meanwhile, my mother payed $50 in 1987 for a blender we still use. It hasn’t come near to reaching the end of its lifespan, but even if it died tomorrow she would have paid $2.27 per year for her blender privileges. And if I wanted to go out right now and spend $50 on a blender, I might be able to find one that would last me an entire year, maybe even 18 months. But I would be extremely hard pressed to find a blender that would last for 22 years.

There are advantages to the recent emphasis on keeping costs low. As a freshman in college, I could afford most of a kitchen’s worth of small appliances. But if I add up all of the small appliances I’ve bought since then, it comes to three toasters, eight crockpots/rice cookers/etc., five blenders, three hotplates, and one electric fondue pot. Nineteen appliances in seven years. One of the rice cookers was admittedly my fault, because I used it for making candles and the wax got into the electrical workings, and the fondue pot didn’t so much break as sit on my shelf unused until I eventually gave it away, but the other seventeen all stopped working. That’s nineteen appliances’ worth of labor wasted, and nineteen appliances’ worth of materials sitting in landfills. I would gladly have traded all nineteen for a single reliable blender and hotplate.

The strange thing is that while it’s a trend nearly everyone’s noticed, no one quite knows how it happened. Over time as products began to be mass produced primarily in other countries, goods became cheaper and cheaper in both price and quality. It’s interesting to think how things might have evolved differently if the manufacturers had just come out one day and pitched this new approach to appliances. “Gather around, ladies and gentlemen, and see the amazing semi-disposable blender! It costs $20 instead of $50, but you’ll need to buy another one in about eight months!” Something tells me that it might not have been quite as popular with the public.

What is the solution to this problem? How, as we’re building a new economic system from the ground up anyway, can we ensure that good craftsmanship is rewarded and customers are not sucked into a wasteful cycle of cheap but shoddy products? It’s actually simpler than you might think, and it’s based on a model that already exists in our current economy: lightbulb manufacturors. When lightbulbs are labeled for sale, they include an expected bulb lifetime in hours. While most bulbs last beyond that lifetime, and a few defenctive ones might burn out much more quickly, it is an estimate of how much use the consumer might be expected to get out of that lightbulb.

What if we applied similar standards to all of our other consumer products? When a new factory was set up to manufacture blenders, for example, part of the process in setting up the factory would be to include an estimated use lifetime for the product. This would not apply to Bill’s ear swab factory, because that is clearly a disposable product, but for non-disposable factory made products, the first few products should be tested for longevity and that information should be included in the Market listings for all future products made with that process. That way customers can make an informed choice between a cheaper product with a shorter expected lifespan and a better-made but more expensive one. And lest manufacturing designers be tempted to cheat and somehow get a better rating than their product deserves, the expected use time also serves as the warranty period. If a product fails before the expected use life expires, it must be replaced at the personal cost of the manufacturer’s Consumption score.

That deserves to be covered in a bit more detail. When the quality of a given product is found to be wanting, just how is fault for that determined? Personal responsibility is important, but so is making sure that the right person is being held responsible. To that end, we must once more interact with our good friend, the jury system. Let’s take the example of one of my many dead blenders. If I was living in a technosocialist society and a blender that was rated to last for 10 years failed after six months, how do we find the cause? A combination of statistical analysis and human investigation is likely to be required.

To start with, we can look at what percentage of units have failed prematurely. The early failure of a single unit out of thousands in use would point towards a failure on the part of the factory worker who made that individual unit. The failure of 10% of the units would point fairly clearly to a design flaw. With that in mind, the fault is initially assumed to lie with the worker in question for products with a failure rate of less than 1%, and with the designer if the failure rate is greater than 1%. This can be adjusted if the worker or designer has unusually high failure rates in other products they have been involved in, or if there are other statistical anomolies. In addition, the party who recieves initial fault can appeal that if they feel they were not at fault, in which case their claims will be reviewed by a six-member panel. Who should compose that panel is a little difficult to determine, because it’s important to have as much expertise as possible without introducing bias one way or the other. My inclination is to recommend one factory worker, one manufacturing designer, one statistician, one engineer, and two members of the general population. A two-thirds majority vote would be required to assign fault in any particular direction, and hopefully the mix of backgrounds would enable the members of the jury to between them make an informed decision. If two consecutive juries fail to reach a verdict, the cost would be evenly divided among all parties involved.

While we’re on the topic, let’s take a second to discuss general allocation of reward and penealty as it applies to manufacturing design. Although the myth of the lone inventor is a beguiling one, it is frequently not all that accurate. Most inventions are made by teams and even groups of teams of people. How can we ensure that they are all compensated appropriately? Basically, this decision must be made within the teams themselves. If half a dozen people collaborate to write a software program, for example, it is up to the six of them to apportion credit appropriately. If two of them wrote the bulk of the code, three of them wrote a few subclasses each, and one of them wrote a single function, it is up to them to decide how to divide the pie. If they cannot decide on their own, a three-person panel drawn randomly from the pool of all people in their field can be brought in to apportion credit. Either way, if someone has done 25% of the work involved in designing a product, they should recieve 25% of the Contribution credit due to the designer of that product, as well as 25% of the liability for any design flaws that might exist in that product.

What of the case of severe design flaws that go beyond simple device failure and cause physical damage or even injury? These would be treated as crimes and dealt with according to the proceedures outlined in the essay on law enforcement, with the sentance applying to those whom the jury found to have been involved in the relevant part of the design process. To go back to the blender example, if the motor of a particular model of blender was found to catch fire sometimes, the person who designed the button assembly on the front would have nothing to fear unless the button assembly somehow contributed to the failure. But the designer of the flawed motor would be responsible not only for the cost of recalling all of the faulty blenders, but for any damage or injury that they blenders-a-flambe caused before the recall.

One of the biggest problems with our current marketplace is that short of a class-action lawsuit, once a customer has bought a product the link between product and manufacturor is severed. The manufacturer does not need the customer to like the product, only to buy it. While most companies would prefer that their customers like their products for the sake of word-of-mouth advertising, one only needs to look at the history of Microsoft Windows to see what a clever and well-funded marketing department can do to ensure a reputation for reliability and ease of use where none exists. By making product quality over time more transparent and making the designers and producers of products directly responsibility for product reliability, the technosocialist model attempts to increase overall product quality while maintaining efficiency of production. As always, the more information available to customers and manufacturors alike, the better their unrestricted choices can help society reach those goals.