I see we also made the CNN crawler at the bottom of the screen.
Detroit Free Press
Dow Chemical plans to cut undisclosed number of management jobs
February 28, 2004, 11:16 AM
MIDLAND, Mich. (AP) -- Dow Chemical Co. is planning an undisclosed number of corporate management-level job cuts as part of an effort to reduce bureaucracy and streamline its organizational structure.
The cuts are expected to be made in the first half of 2004, said Leslie Hatfield, a spokeswoman for the Midland-based chemical giant.
"The chemical industry is under intense competitive pressure," Hatfield told The Saginaw News for a Saturday story. "One of the measures we've taken is corporate restructuring to consolidate some jobs."
The changes were expected to reduce Dow's cost of doing business and condense the company's day-to-day operations.
In January, Dow said it swung to a fourth-quarter profit, despite higher raw material costs, because of increased demand for its wide array of products. A year earlier, the maker of chemicals, plastics and fertilizer was hindered by merger-related costs and asbestos-related charges.
In 2003, the company reduced structural costs by $600 million and improved its cash flow by more than $2 million compared to 2002.
After dismal results in 2002, Dow announced it was cutting from 3,000 to 4,000 jobs and selling or trading up to $1.5 billion in assets. The company now employs about 46,000 people worldwide.
"After his presentation he was told that he "could have won first prize if he'd mentioned cost savings". These are not nice people! "
I'm surprised they didn't tell him that now he could neither enter heaven nor reach nirvana.
Those course notes were written by a "team".
"Teams" can do no wrong, as you have learned.
When I said the notes were wrong, I meant wrong in the strongest sense of the word. Wrong in concept, wrong equations, wrong conclusions, wrong clear to the bone. Also, the instructor notes were not coordinated with the student notes; some pages in the student notes were different at one GE site than at another. My attempts at fixing the notes in public (while teaching) drew scorn from the students.
It's not "everything was wrong". They had a MBB at one site who saw clearly what was happening... complained to the fount of all knowledge at their Mass. HQ site... but he was ignored. He and I both knew the notes were junk, but he was stuck in the game and I was not.
Let's go a little further. In another company (I wod work for them) their prime customer is GE. They were forced to do a Sick-Sigma project with GE. This project didn't bear fruit. In spite of this GE invited their team leader to present a "success story" in Pittsfield Mass (GE Plastics HQ site) to an audience that included team leaders from other GE customers. The team leader was forced to present "something". At the last minute the GE sponsors told him that he should mention and play up the cost savings. There were cost savings... the results of the project could not be reduced to practice. They tried to strong-arm him into telling about the "cost savings". His manager realized that if they even mentioned $ savings then GE would want to "gain share" with them. So he didn't mention cost savings. (It would have been a lie if he had.) After his presentation he was told that he "could have won first prize if he'd mentioned cost savings". These are not nice people!
I quit my last job because our company adopted the GE style "Sick-Sigma".
You can't say the emperor has no close under Sick-Sigma because, by definition, all work is essentially flawless.
So how could there be 17 pages of incorrect stuff in the notes? Wouldn't that violate GE's six sigma goals? Don't they practice what they teach?
I could write a book on the "belts of many colors". About 6 months after I left UK I signed a contract to teach Sick-Sigma for GE Plastics. That was the worst mistake I've made so far. That was one of the worst experiences of my career. I was in charge of teaching budding BBs and MBBs. The course notes I was given to teach from were a total disaster. In one instance I eliminated 17 pages of stuff that was just incorrect and substituted three pages "done right". I'd gotten tired of correcting the notes while teaching, so I attacked the problem in one gulp. The belts-in-waiting howled and complained because I'd changed the notes. They said "those notes were written by GE and you have no right to change them." I knew right then we had a problem. I quit... broke the contract... a few sessions later and basically told them to go pound sand. So far, GE is one of two clients I've "fired". The other company makes kitty litter.
Here's a major problem with the whole Sick-Sigma thing. All improvements are attributed to Sick-Sigma. If the work is done by a single isolated individual, that doesn't count.... it's got to be a "team effort" and that team must be a Sick-Sigma team.
Many BBs and MBBs think that Sick-Sigma is their future, and believe that there is a large market for their services. Not so!!! I get about one e-mail a week from BBs and MBBs who have been "let go" and they are learning that the world is awash in belts of many colors.
I still teach 6-Sigma for some companies with the understanding that there will be no "belts" and we'll find a better name for it.
The methods I cited will tell you the amount of statistical bias. If the calculated statistical bias is the same as what you are seeing in the plant material balance, then it's a good bet that the material balance offset is due to the statistical bias.
These statistical biases are sneaky rascals. They spawn simply from the way calculations are done, pure and simple. I've seen good people struggle with gaps in material balances and other things (like the stack measurements) and struggle and struggle without realizing that they are dealing with statistical bias. This kind of bias is known, but it's not well-documented. I've never seen reference to it in an engineering book. It's nicely buried in higher-level statistics books... not in Stats. 101.
I got into some new aspects of this (stack testing) by working on a contract for the ASME. The ASME has very specific formats for evaluating the performance of machinery... enery efficiency, etc. Their formats (recipes) are very specific, hard-core, and the calculations take into account every correction you could imagine. But... I've noticed... the nature of those calculations opens the door to statistical bias by definition. Pointing that out to the guys who run the ASME was like bringing a snake to Sunday School. I've run some Monte Carlos to show that these biases are real (albeit, not large in that context) and those have gained some attention.
Don't let Engelhardt off the hook. I was down there on a catalyst recovery monitoring trip while they were running around looking for a 55 gallon drum of platinum. After two week, and three different acid spills that holed us up in the control room, the best they could figure was that it must have been shipped out with someone elses catalyst. I don't know if they ever solved that one.
I do know it did not go out in anyones pockets, without guard collusion. Their detectors were so good that they picked up the foil on a pack of rolaids I had in my pocket.
Before I left, I sat in on a discussion reviewing a BB project. Our ignorance was being highlighted to those who would remain behind. The project was taking credit for a "cost reduction" as UK was using a higher quality (costlier) raw material for a product than called for by the product specs. The leader got a bit annoyed when someone pointed out that the same RM was also used for two other products and the volume discount be received for the larger volume offset most of the added cost for the item under discussion and added points with the customer.
After I left, I learned another BB took credit for NOT changing that product spec and used the volume discount as the rationale for a savings by staying with the heritage arrangement. Two different proposals taking credit for cost savings when nothing was changed. For a while, I thought John and Ruth were "working" for Dow.
I guess we were just too dumb to know better.
The amount of bias is predictable. There are equations (approximations) for estimating the amount of bias. For complex systems that route is a mess. I use Monte Carlo to estimate the amount of bias. Sometimes there are complications with that route, but it's usually good for a quick perspective.
The stack sampling situation is a cute one.
Convincing the EPA that there's a bias, and documenting that "once and for all time" was an enormous undertaking.