Saturday, May 21, 2005

Twelve layers of management

From Anonymous:

I’ve been trying to figure out where I fit in the grand scheme of things and went to the LANL web page to try to figure out the management hierarchy. Here’s what I came up with:

Title (UC - LANL)

Title (Real World)


Laboratory Director



Deputy Director

Vice President 5


Principal Associate Director

Vice President 4


Principal Deputy Associate Director

Vice President 3


Principal Associate Deputy Director

Vice President 3


Associate Director

Vice President 2


Deputy Associate Director

Vice President 1


Division Leader

Division Manager


Deputy Division Leader

Assistant Division Manager


Group Leader

Section Head


Deputy Group Leader

Assistant Section Head


Team Leader

Group Leader




Does anybody really believe that having twelve layers of management between the scientist/engineer in the lab and the boss is productive? .

Notice Titles 4 and 5: is there any meaningful difference between them?

Management theory and experience suggests that the optimal number of direct reports for a supervisor is between five and ten, with eight being a generally accepted number. If we do the math (log(base8) of 13,000 employees), we get 4.6, or approximately five layers of management required. Do you wonder where the overhead goes? Do we really need five levels of vice presidents?

Or are we like the old Ma Bell, where, as a cost plus business, the bureaucracy grew like Kudzu to kill the organization under it.

Add one more layer for a total of 14 (3X a many as needed if you had good managers), you forgot the position Associate Deputy Director the one M. Devaurs has and that M. Burns had before her.
Hey, you left out program managers.
The management problems in a large part are a consequence of Sigfried Hecker's "flattening" of the management structure by tripling the number of managers. The consequences are clear from a bottom-up perspective: no clear lines of authority and no lines of accountability. If a prospective contractor were to propose an organization chart with functions but no names, 75%-80% of the current managers would find themselves declared redundant.
While the point is that we are management top-heavy is well taken, this analysis is highly questionable.

I do not acknowledge all of these management positions as "levels" and it is rare (to the point of literally never happening?) that anything would go through everyone of these hands on the way up or down.

From my lowly position as TSM and sometimes Team Leader, I see GL/DGL as a single level unless there is extreme dysfunction going on. I see DL/DDL the same way. I do not see much of what goes on at the AD level directly, but I don't see all of the Associates and Deputies to the Director as an extra level. Oftentimes, I don't see team leaders as being a real level of management, so at worst, it seems like:


5 levels and maybe 4 if you discount TL roles

Where I work, TL's work on projects full time and do whatever TL duties they have as a matter of course, very little more than senior staff with mentoring and advisory roles. DGLs also work 50% on programs and GLs often as well.

Log8(13,000) = 4.6 .

I think someone already did this math and things are nominally set up that way.

I do not feel the oppression of too many "layers" of management. I feel the oppression of the individuals we have in those layers being upwardly focused, turning downward only when they need something (like a whipping boy or scapegoat or some things to take credit for).

I talk directly to my GL/DGL fairly comfortably. I would like to be more comfortable with my DL/DDL. And it would be nice if I didn't feel completely at risk talking to my AD. I'm not talking about short-circuiting chain-of-of command, I'm talking about "getting with the program". To get with the program, we need to understand the program and possibly to influence it...

Corporations and the military are usually very top-down. For many reasons they need to be. Can a scientific laboratory be as strictly heirarchical as the military or a corporation? I don't think so. A scientific laboratory is a microcosm of a scientific community. There is more need for and should be more room for communication and respect across all lines of the heirarchy.
The previous comment is right on. Many of these listed do not have direct reports who are also managers/supervisors (e.g. all of the 'deputies', and so they don't count as management layers. If you counted this way, the military would have 30-40 layers. Organizational structure is often blamed (and changed fruitlessly) when incompetence is really the culprit.
When I started work at LASL in 1965, the only people between me and the Director were the Group Leader and Division Leader. Deputys served in the absence of the GL or DL, period. The beginning of the current organizational nightmare started under Don Kerr's watch. I think he wanted the Lab structure to mirror the DOE.
I have worked with very good management who both maintain "chain of command" and an "open door policy" at the same time.

I'm not sure it is that hard, but it does seem rare.
How can anyone try to explain with a
straight face the title
_ "Principal Associate Deputy Director"?
Here is one only slightly less
absurd that I had on my door before
quitting the Lab a few months ago:
_ Assistant Principal Associate Vice-Deputy for the World's Greatest
Science, Acting and Soon-to-be-Ex-TSM.
_ Has a nice ring, doesn't it?
Anonymous : 5/21/2005 10:49:54 PM said:

"The beginning of the current organizational nightmare started under Don Kerr's watch. I think he wanted the Lab structure to mirror the DOE."

I suspect, if the truth be known, DOE forced the structure on Kerr just as they tried to force it on Agnew. If Harold could not reason with the bureaucrats, Don did not have a chance.
It's good to read these posts. Many of the comments in this blog suggest to me that the writers think that things at LANL went sour only in the last 3-5 years or so, but particularly under Nanos' "Reign of Terror". Not so. I think many of the Lab's problems (poor management, increasing bureaucracy, high overhead costs, etc.) have their roots going back many years. Specifically, I think the mid- to late-70s marks a major turning point ("downhill", actually) at LANL. Nuclear testing declined significantly after this, as did the number of new warhead procurements. To take up the slack, the Lab increased its efforts to broaden out into other scientific and technical fields, even though the Lab was simply not "world class" in many of these (and still isn't). The number of "reimbursable" programs increased, and LANL slowly began to look more and more like a "Beltway Bandit" contractor. When DOE came along, it marked the death of any remnants of the AEC. The DOE, other than manage the weapons complex, did not, and still does not, have any other reasons to exist. It did such a crummy job with the labs over 20+ years, that we got the NNSA, which is unfortunately no solution either. In the interim, the federal government continued to grow in size and budget, and in so doing, greatly increased its already huge bureaucracy, and form trimphed over substance. These factors caused LANL to do likewise because it needed new programs in addition to the declining weapons work. LANL Directors after Harold Agnew started to look a whole lot like the high-ranking, influential Washingon, DC civil servant politicos who couldn't pour water out of a boot if the instructions were stamped on the heel. (John Browne was a reasonable exception, IMHO) The number of managers at all levels proliferated, and lots of new levels were born. I think Washington encouraged, and loved, it. It showed them we wanted to be just like them. And now we have oodles and oodles of "assistants", "deputies", "principals", "chiefs-of-staff", ad nauseum that didn't exist way back because these positions had no valid reason to exist. And, Good Lord, then there are the myriad program managers, many of whom just live off their subordinate TSMs' and Techs' hard work and add no value. Nanos was the icing on the cake, but not the original instigator. But I hope he marks the evolutionary dead-end of what's gone on the last 30 years or so. My "analysis" here is admittedly simplistic, but I think there is something of a germ of truth in it. The next couple of years will be very interesting.....
The 9:53 posting seems to have the ring of truth. Many of us were not around in the period he describes. I would like to hear more about the scientific bases and outcomes of the "programs" he observed at LANL? Some posters have described them as flops and embarrassments. How could that be true? How about Browne's program called ATP? Was it successful? Does it have a future? Did our public relations portray it properly?
I was around from 1978-2003 and poster 5/23/2005 09:53:20 AM is right on the money. Upper management lost interest in nuclear weapons even faster than the public at large. They also forgot that the best ideas come from the working level scientists. They lived in hope of a rebirth of LANSCE which every outside expert said had reached the end of its useful life. Or they tried to dream up new replacements for LANSCE. They made up projects like ANTARES where they had conceptual drawings showing where the toilets would be long before anyone understood the interaction of laser light with materials well enought to choose the proper laser.

In the meantime many really good basic physics capabilities were allowed to wither. Los Alamos had the best radiochemistry group in the world and John Emille desided it was too expensive to keep. They were capable of doing important envronmental work as well as non-proliferation work and direct weapons and testing work, bye bye.

The BEAR program was very successful in terms of the staff accomlishing everything they were asked to do, but when they got there there was no there there.

And every year the Lab brought in some expert like Motorola to study the Lab. Every year they reported the same thing: the lab has great staff but the management stinks. And every year the Directors response was "oh I see we need more floggings".
This is the 9:53 poster back again. A previous poster asked about the Lab's "flops" over the last two or decades. First, I think the weapons-related organizations in the Lab have done quite well, all-in-all, since the Manhattan Project days. I recall being very impressed when I was involved in my first nuclear weapon shot in the early 1980s. People in all parts of the project, from weapon designers at LANL to the diagnostic physics groups, to the old J-Division, to the guys who drilled the holes in Nevada were serious, smart, hardworking folks. Missing a shot deadline was not a mark of accomplishment, and no one wanted to be the guy on the hot seat if he caused a snafu with the shot. Ditto for the other weapon-related organizations at LANL. And there have been many other successful projects, smaller in scale than the weapons projects, and in other technical and scientific areas. Some of these are known, others are not known (often because these were successful and so became classified). Overall, I think the Lab has had many successful projects in the last 20 or 30 years. There are also other projects that were successful technically at various levels, but the government sponsors pulled the rugs out from under them. Examples are METEOR (an SDI funded laser project) and the FEL program (poorly managed, but had great technical strength at the worker-bee level). LANL actually beat out LLNL to get the contract for the SDI FEL testbed to be built at White Sands! However, there have certainly been some major flops. For example, the AURORA project, an excimer laser-based ICF testbed that followed ANTARES, was a management debacle. The TSMs and Techs didn't stand a chance. Another flop was the outrageous expenditure on environmental projects that sprang up in the early 1990s. There were lots of managers of all kinds with money to do "projects" and "programs". They talked a lot, generated tons of paperwork, and took factfinding trips to Russia and elsewhere, and yet essentially accomplished no useful work. But, Washington and DOE apparently thought something good was going on, cause the money kept flowing. I could easily go on and on, but I'll spare you. My overall observation is that the most successful projects at LANL have been those where the technical people from LANL and the sponsors are allowed to get together, with minimal armchair quarterbacking from the higher level managers. But, when we have way too many superfluous managers who have to find something to do to justify their existences by shoving a project down their subordinates throats, the odds of a failure go up. This has happened on more than a few occasions within the last 10-15 years. Another big problem with scoring a success these days is that the overhead costs at LANL are so big, we find it difficult to get enough money to actually do the project. When I joined the Lab, the entire budget was less than $1 billion. Now, it's $2.2 billion. Certainly, legitimate costs over the years have risen, but not by a factor of 2.2. A lot is the increased overhead to support outreach centers, diversity offices, safety and security training (and the people that go with this), and the "principal associate deputy vice chief-of-staff" and the like in the Division and higher levels. I recall Sig Hecker claiming several times he wanted to eliminate "checkers checking checkers" that plagued the Lab. He didn't succeed, in part because he helped usher some of them in (or was forced to by DOE and politics).
If by "outrageous expenditure on environmental projects that sprang up in the early 1990s" you are referring to GEONET, I take exception. GEONET was well on its way to great success, and the expenditure were minimal. It was axed at the knees by management. Argueing about which project was good and which was bad,however, is sort of beside the point. The point is that upper management was directing all of the programs and their track record was abyssmall. The Lab needs to have a staff empowered to bring forth new ideas.
Watch out! We’re opening closets and exposing skeletons. I believe that some of the posters do not fully understand what they are describing. The programs that are mentioned are all dead with no chance of resurrection. There were no "lessons learned", either. Try and find a written reason for the various collapses. Try to find how the programs began!
These, and quite a few others, were handed to LANL staff without the benefit of peer review or, even, intelligent planning. Remember the giant generator that came up the hill? It wasn't ever turned on!
These projects formed today's LANL, not Nanos. From the point of view of an outsider or one who left the lab in horror, the policy that led to these debacles must be changed.
5/23 4:35pm here again. No, I don't include GEONET as one of the environmental program flops. Instead, it is an example of having the rug pulled out from under it. As for the statement that we may not understand what we are describing here, I worked on most of these failures I mentioned, so I know a lot about how these came to be, and why these bit the dust. To me, these were projects that were killed by inadequate technical and personnel management, with healthy doses of bad politics and overinflated management egos thrown in for good measure. I don't think there's any reason to keep the skeletons in the closets. If it is not classified, people should know how their tax dollars were wasted. And perhaps there are classified projects that were made so and kept that way just so few would ever know about the waste. Our federal government has done this lots of times.
Post a Comment

<< Home

This page is powered by Blogger. Isn't yours?