Saturday, May 14, 2005

More advice for the bidders

From Anonymous:

A couple of years ago, when I still believed Pete Nanos had LANL’s best interests at heart, I sent him the following advice on LANL culture and management. The advice seems to be still applicable today. Perhaps the winning bidder will take this to heart.

Management areas that must be improved:

1. The absence of a strong program management function that provides third-party over-sight, reporting, and corrective actions for work packages that are in the early stages of trouble. The LANL program management function focuses on the beans of cost and schedule without addressing technical issues. Systems engineering is non-existent. This results in crisis management as technical problems fester and then finally bubble to the surface when they are very expensive to resolve.

2. The absence of a structured management development program to give people promoted from the technical ranks the “people tools” they need to do their jobs. There also does not appear to be a management succession plan to identify, develop, and promote well-qualified leaders into the management ranks. There is apparently no mechanism to identify under-performing managers and to offer them a face-saving exit when they don’t work out.

3. A culture of “collegiality” which is analogous to the Military Academy “Ring knockers.” People are reluctant to criticize their fellow workers because of the strong negative tribal consequences and risk of professional ostracism. This builds a “go along to get along” mentality where preserving the UC contract is the ultimate goal and hiding problems is the cultural response. You are not allowed to criticize someone who screws up. I don’t think you can build a culture of continuous improvement and customer focus in this climate.

4. A policy of hiring new grads and post docs with little or no industrial experience. As LANL becomes more involved in big iron engineering projects, people with industrial experience are more and more crucial to success. Hiring inexperienced people means you get to buy their learning curve.

Advice from the trenches:

1. LANL needs to manage their managers, rewarding those who are capable and moving aside those who are not. You need a face-saving mechanism to replace under-performing managers just as you replace under-performing TSMs.

2. Team Leaders and Work Package Leaders (first level managers) need more support and training. The senior management culture must respect their contribution while simultaneously providing the mentoring and oversight they need to keep out of trouble. Project management that focuses only on cost and schedule to the neglect of technical performance is a two-legged stool.

3. LANL needs to build a culture of engineering discipline. There is no apparent attempt at imposing the disciplines of systems engineering, formal specifications, peer reviews, or test and validation plans on product development. Part of it is the LANL prima donna mentality, and part of it is the emphasis on hiring people who have little or no real world experience (new grads and post docs). Emphasizing new grads and post-docs means you are always paying for the learning curve.

There are some good suggestions in here, however do you really think the country is best served with another engineering laboratory? I think there needs to be strong support of science, that is not encouraged by Gant charts.
As the original poster, I was noting that much of our "Big Science" requires advanced engineering that we do poorly, with DARHT second axis, the leaking SNS cavities, and exploding magnets as good examples of projects that spun out of control due to lack of engineering discipline.

It's not a choice between being a science lab and being an engineering lab - it's a using a disciplined approach to the engineering to facilitate the science and keep us from becoming another NIF.
It's funny, but a couple of years ago, I too believed that Pete4 Nanos was good for LANL, and I too sent in several "TellPete" messages with positive recommendations. Here are a couple of them (from memory).

Get to an "effective" safety and security program. LANL does these for the sake of doing them, in my opinion, and they are way overblown and burdensome. Somehow quality assurance needs to be in the same sentence, i.e. safety, security AND QUALITY need to be emphasized. You can't have S&S without Q. Effective means just enough to get the job done. Give us requirements, not instruction manuals. The technical staff should figure out the "how" while the ISSM folks give us the "what", "when" and "where" of compliance.

Another major LANL problem is discrimination, fixing this would give LANL a major boost. The salaries for women are well below those for men in comparable jobs, but not much different from the national average, so LANL isn't under performing in the salary sexual discrimination category. Where they really have a problem is in the area of equal opportunity for advancement. It is extremely hard for a woman to get promoted to the higher level jobs at LANL. Especially for technical women, there appears to be a feeling that women have no place in upper management in a physics laboratory!

The second major discrimination area is that of age. Anyone who has more than 2-3 years at LANL is underpaid compared to new recruits. If you have been here 5 years or more, the difference is dramatic. Did the Welch analysis look at this aspect at all?? I understand it was considered, but didn't count in the overall "points". It appears that even if you do have experience in industry, you will not be compensated for it, so why would you want to come to LANL? Those of us who do have industry experience are looked down upon for it by the prima donna scientists who know only graduate school and the brotherhood of university "connections". There are many industry scientists who have PhDs in the "school of hard knocks" but only "the paper" counts to LANL when hiring and promoting. This attitude eliminates a large number of very fine American candidates who decided to work instead of continuing on in school.
8:07 is correct that "there needs to be strong support of science" and that the path toward that goal looks different from what the original poster described. The fact is, however, that there are parts of the lab that make their living off of engineering rather than science. The original prescription may (or may not) apply well to those parts.

Not only bidders but Bob Kuckuck and his inner circle would be well advised to accept that, far from a single "laboratory culture" that needs to be "turned around" as a certain unnamed but recently departed manager believed, there are multiple cultures here, and they don't all have the same needs. One of the things Doug Beason (there's that name again...) did right in his first all-hands session with TR was to acknowledge that all four of his divisions had different challenges to overcome if they were to prosper. He then set about doing things to help each division on an individual basis with their individual needs. It's too early to tell whether his specific remedies are going to work; my gut feeling is that some will, some won't (and one may actually make matters worse). But he's on the right track.

Before implementing the original poster's suggestions or any others, I think it would be worthwhile to take a detailed look at our needs and develop some individualized plans for dealing with them, probably at the division level and at least at the directorate level. This should be done by some fact finders who don't just listen to canned presentations but really listen to people as they try to do their jobs.

The division review committees are theoretically supposed to do this, but it doesn't work, because divisions have powerful incentives to tell the DRCs stuff to convince them that everything in the division is rosy. Furthermore, the DRCs tend to be politically naive. I don't know how many times I've seen our DRC put out recommendations that the lab should persuade Washington to support program XYZ -- as if we can just snap our fingers and make Washington change its mind. The result, at least in our division's case, is a DRC that is unquestionably well intentioned but completely otherworldly in their view of what's going on. That doesn't help.

Rather, I think Kuckuck should try to do this via *internal* review conducted by people whose job is to identify needs, not to either make glowing reports or throw stones. This needs to be done in a no-fault way, because it will only be productive if those interviewed by the reviewers can speak freely without fear of getting screwed. Personality conflicts should be kept out of it, and input that does nothing more than whine about leader X or policy Y should be downplayed, although not necessarily disregarded. The goal isn't to replace leaders or policies; it's to determine what's keeping us from succeeding at our jobs and propose some fixes.
To 8:58's Beason point:
I don't necessarily agree with your assessement of Beason, perhaps he started correctly, but he is off the path now. I have heard the following comments taken from another comment post which leads to my disagreement:

At first those of us in ADTR felt this was a wise choice from a leadership perspective (he made positive changes in ISR) and certainly a step-up in terms of personality (to be mild) from Cobb (unfortunate for the rest of the lab with his current position). However given current decisions we in ADTR have begun to wonder if Beason has become disconnected (perhaps not concerned) or beyond his leadership abilities. In particular his choice of: 1) morale officer, 2) keeping N-div leader (who has destroyed N-div & continues a crusade of intimidation & reprisal, e.g., N-3), and 3) the ISR div. management selection sham (choosing the ISR Chf. Scientist for a div. leadership position, that individual is distrusted & disliked by a majority of staff in ISR, has a poor record of institutional & inter-divisional behavior, has no people skills nor line experience, has a questionable security history which sends the wrong msg. to staff, is the epitomy of self-interest, and has a knack for blithely claiming ownership of others ideas).

How does the lab choose its leadership/mgt. personnel?
Regarding the comments from Anonymous : 5/14/2005 08:52:07 AM, the laboratory now has a committee working on "Conduct of Engineering." Unforetunately, many of the individuals on that committee were resonsible for our engineering failures on major projects. So, you can imagine what the expected outcome of the committee's work will be.
Was Priedhorsky really promoted to the ISR division leader job? Is that a fact, rumor, or pending UC confirmation? Was it announced? If so, that is totally amazing. I have never heard him claim anyone else's ideas, but I know he has zero leadership experience. If true, this is strike 3 - after Rao and Lisa!
To Anonymous @ 5/14/2005 06:29:36 PM:
What are these people thinking?
There are completely different needs in different parts of LANL; God save us from universal solutions like "hire only postdocs with industrial experience", or "hire only postdocs who are specialists on relationship of M-theory with loop quantum gravity". There should be hired the best people for the job, and usually not narrow specialists, who can learn other things.
The sentence:
"There should be hired the best people for the job, and usually not narrow specialists, who can learn other things."
should read of course
There should be hired the best people for the job, who can learn other things, rather than narrow specialists.
Post a Comment

<< Home

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