Wednesday, August 15, 2007
Scaling
Post contents to be reposted.
Comments
-
and ...
the bigger you are, the deeper you go in all these calculations.
I’ve always worked on the INSIDE support - managing the materials, supply, and services necessary to keep manufacturing going. All the information about inventory gets even worse when you consider inventory of spare parts and materials (MRO) necessary to keep the machines working. Those machines that are aging (as are the workforce) need emergency repairs, planned maintenance, and upgrades. There’s always parts that are vital ... if one breaks, the machine and probably the whole line is down. We deem those “critical spares” and have to buy and keep them on a shelf .. just in case. And keeping spare parts invesntory is as much (pr more) of a cost as keeping sales inventory. Their stuff moves ... mine doesn’t. So I have to pay taxes, heat the building, employ warehouse staff, pay insurance, etc. It costs approximately 35% of the inventory value, each year, to maintain that stuff.
We not only have to do ROI caculations on new equipment/lines/plants, but also on how much new inventory we’re going to have to support.
Economics is fun ... layers upon layers upon layers.pete in Midland | 8/15/2007 02:18 PM CDT -
Hm. I have been in the business- these past fifteen years- of increasing the number of hours in the day, and I’ve been doing it successfuly all that time. So what’s the problem? Maybe all this is just something I’ve always taken for granted. If you can make 1000 widgets a day, and you need to make 2000 widgets a day, your day just has to have 48 hours. I do that all the time.
og | 8/16/2007 08:34 PM CDT -
You may be in an unusual situation. You actually like what you do. Most of us don’t like what we do for a living so the encroachment extends into the continually reduced time of our life that actually is pleasant.
No one minds (or should mind) the occasional week or month where 2000 widgets are needed. It is when it becomes the new baseline, and every week is an increasing demand.
Mrs. du Toit | 8/16/2007 09:04 PM CDT -
This brings to mind the whole minimum wage debacle.
For instance you’ve got good ol’ Bob who’s made widgets for years. He can put out 100 widgets an hour. So you pay him 10 cents a widget or $10 an hour.
On the other hand you have the new guy. He can only do 25 an hour. So you pay him $2.50 an hour, wrong, the law says you have to pay him $5 an hour. Labor cost has gone up 50% for a 25% increase in production.
So what do you do, cross your fingers and hope your customers will pay more for genuine made in USA widgets or look for a Chinese company that will make quality widgets cheap?
It used to be accepted that when you first started a job you were going to be paid starvation wages until you made your labor profitable enough for the company to pay you more.
I can’t blame a lot of companies for outsourcing their labor overseas.
rlhunter | 8/16/2007 10:14 PM CDT -
No, actually, if it takes a machine 24 hours to make 1000 widgets, the laws of physics are pretty much inviolate. You add a machine. That gives you 48 machine-hours every day.
These days- modern machinery being what it is, reliability is not an issue. A factory robot has an average of 72,000 hours meantime between failure. If a robot can- for instance-polish 250 pistol grips a day, two robots can do 500, four 1000, etc. That’s what I call putting more hours in a day. Scaling? that’s what I do to fish. You need extra machine- hours you need more machines. And the roi on robotics- in certain industries, anyway, is about 8 months. Great post- and I like all the descriptions. The type of stuff I do bends those laws in a lot of ways, so most of those descriptions have changed. What you say, though, is absolutely correct: there are physical limits.
og | 8/16/2007 10:31 PM CDT -
Which is why the next section talks about ROI. 8 months assumes the market increases… which are not always a given.
Mrs. du Toit | 8/16/2007 11:50 PM CDT -
Yeah, the way we sell it is that ROI should be covered by existing orders. In other words, if you have a contract for 1,000,000 widgets and that will take nine months to make, only buy equipment that can be paid for in that nine months. At the end of that, even if you lose the contract, you still have the equipment. We also usually sell 25% productivity increase with each system so the ROI figures we generate are always very conservative.
Yes, sometimes contracts do fall through- but you can’t be chicken little all the time. The difference between what I do and what you do is enormous- I can provide machinery as fast as people can practically expect to use it- you have absolutely finite resources and once you have plumbed the depths you’re basically done. You can do no more than you can do. I’ve been thinking of ways to apply my skills (automation) to your type of business. I don’t think there is any way. I’ll always be a nuts and bolts guy, at the end of the day- I like pulling something out of my head and seeing it in the real world, a piece of machinery or an object you can hold in your hand and touch. It’s how I’m wired.
og | 8/17/2007 07:58 AM CDT -
Having spent the majority of my working career in non-manufacturing, finding efficiencies has become increasingly difficult.
In fact, it has been my experience that we reached a point of over-automation because folks are thinking that any automation will increase productivity (the mindset of it did before, why not again?).
In the more recent cases the automation has become something like bureaucratic red tape.
We’ve reached the point of silliness, where folks are insisting that a 21” monitor (instead of the 15” one) will enable them to do their job better. Crazy! It is just prettier, but no productivity increases will be found doing that. If the person is getting older and needs bigger screen print, that’s different. That is a maintenance of current levels issue, not an increase.
Mrs. du Toit | 8/17/2007 10:39 AM CDT -
There is also an ROI on change management that few factor into their plans. During the period of transition productivity will be less, and it is often the case that it remains less until the system has been tweaked and tweaked. In my experience, in MOST cases, the benefits of automation are never found, in the form of productivity increases (which is a method of lowering costs).
There may be benefits besides that, but few go into the process with the expectation that it won’t save them money.
Mrs. du Toit | 8/17/2007 10:42 AM CDT -
You do have to know how and when to use automation. That’s my gig. I’m extraordiarily good at it.
og | 8/17/2007 06:42 PM CDT -
I am fascinated by the fact that you see automation and think manufacturing. I see it and think service process. Both exist of course, but my main experience (because of the nature of my work) is an automating a process that people perform, such as producing reports or paying bills. Totally different animal than the manufacture of a thing.
The aspect I work with is back office stuff, not production line.
I can reorganize an office to achieve the highest produtivity and work satisfaction possible… it may be achieved partly with automation (the use of computer software to perform redundant tasks), but that is always a small part of it. “Right sizing” is the trick and there is no automation involved in that.
Mrs. du Toit | 8/17/2007 11:21 PM CDT
