Wednesday, July 27, 2005

Why we are not as productive as we could be

Hi Doug
Could you post this anonymously ?

Could you post this page 6 from IMP 352.0, Implementation Procedure on the blog. It is a pdf file
and I don't know how to do this. I would put it as a response to the productivity issue.

I think it is a nice illustration, why we are not as productive as we could be. I really wonder, what a level 8 milestone would mean. And also once you take an average inflation of 4 lower milestone/per one upper we will get 84=4096 milestones a year.
Assuming 240 days in a working year, will give us 17 milestones a day. Further assuming
it takes 20 minutes to write a report fo every milestone, we will be roughly using 6 hours a day
working on these milestones. I wonder if this is the "Foley" business model?

A few personal observations:

1) this is why Fred Tarantino's organization is truly humongous, and truly expensive to the Laboratory. All those program people and their support and administrative staff don't execute much, if any, technical work. They get money, skim off a bunch to run their organization, send what's left down, and wait for paper and powerpoints to come back up.

2) I've learned that if one can't map their work to a level two, or at worst, level three milestone, clueless program managers want to know what your project does that's important to the weapons program. When one tells them over and over again, they repeatedly fail to understand and still want to zero it out, even if killing your work would kill the weapons program. Really.

It's entirely analougous to refusing to pay for electricity because it can't be tied to a level two milestone, but that's how it's happening.

3) I have yet to hear anyone ask them to defend what THEY do that's important to the weapons program, let alone ask and then ignore the answer.

And we pay these guys and gals to do this to us. I feel like we're paying for someone to strangle us, and they're asking as we turn blue why stopping is good for THEM.
This is classic portfolio management, and it's done every day in the Real World. The last three levels are obviously reserved for internal project management, allowing flexibility for projects based on size, risk, and complexity.

One should try living the life of a program manager for a while before slapping a Clueless label on everyone with that job title. For small programs, it's one or two upstream clients and a basic work breakdown structure. Not hard to do on the back of an envelope if you're so inclined. Larger programs have commensurately larger client bases, reporting requirements, and scrutiny. That's when life gets fun, especially when you have to balance the needs of a large program with the needs of solo scientists with their own funding mechanisms. This results in a clash of two different worlds.

The big problem is training all the program managers (and program element managers) in classic project management and control (PMAC) techniques. Most program managers started out as technical leaders that hooked up with a steady set of clients. With backgrounds in science+sales, things work OK until the organization tries to implement a scalable, integrated model. Then you get clashes like this because many Program Managers and Project Leaders are on unfamiliar turf.

Either way the contract goes, LANL is going to start operating like a Real Contractor. That will cause problems for those who can't let go of the Good Old Days.
I manage a project with Level 6 milestones reportable, Level 7 internally monitored, no Level 8. L6 milestone reporting takes me about an hour every 4 to 6 weeks, depending on project progress. Not exactly a monstrous burden on the productivity of me and my team. Methinks the original poster needs some remedial math training.
I think the correct number of milestones is more like 4^ even larger number than in the original post. On the last hydrotest, we had more bean counters than it once took to do the whole job. Welcome to the new Los Alamos where genuine productivity has become a lost art.

Scott Watson
"Welcome to the new Los Alamos where genuine productivity has become a lost art."

Welcome to the new Los Alamos, where actual metrics will be applied to quantify productivity drive efficiency. The lost art will be receiving money without ever having to justify what you are doing with it and when it will be done. I am sure the tax payers will be grateful.
I don't need a Gantt chart to manage a project, be productive and deliver results. The Project Management doesn't produce anything, it only assists the researcher, and in many cases, it is impossible to match the "plan" with the resulting "research" as research just doesn't work like that. There are many blind alleys, backing up and starting over, not to mention new ideas and direction changes. I don't mind a certain, appropriate level of PM, unfortunately, DOE has over-engineered this program just like most others at LANL, into a bureaucratic, self-feeding system that will cost 2-3 times what was being wasted before in excellent, creative research. The prospects are mind-numbing, and when that happens, creativity goes out the window. Believe me, I know, I am only creative in the wee hours of the morning lately, at work, the tension shuts down my frontal lobes.
Ok, "anonymous" 6:48 I cry foul. What 'Actual Metrics' can you site that are being applied to Los Alamos hydrotests? The GAO who recently audited the hydrotest program didn't exactly agree with your sentiment that the taxpayer was getting his/her dollar's worth as a result of LANL project management. Their metric was a simple and reasonable one - data delivered on time per dollar. Perhaps if the metric had been number of people employed tracking projects that fail to deliver, they would have come to a different conclusion - one more in line with yours.
- Scott
I have found the general direction in program management taken within the Nuclear Weapons’ Program to be immensely disturbing. I think the sort of bureaucratic bloat that the chart shows is merely symptomatic of the true problem. Rather than gripe about all the milestones, ask the question about whether the program is effective.

Here’s the question, with all the programmatic formality are the programs any more effective? Has Tarantino produced a better program?

Here’s my answer, an unequivocal NO! The answer is because the proverbial tail is wagging the dog; the metrics for the program have taken over. It has become so important to succeed with all the milestones that the program makes the decision to dumb-down the milestones to make them easy. The program has become completely risk-adverse, and high-quality work depends on assuming risk. Risk of failure is essential for quality. I am observing a systematic reduction in the quality of the technical content of the programs I am part of. This is all being driven from above with the milestones and earned value metrics.

Bottom line, there is nothing intrinsically wrong with these metrics if they are applied appropriately and get results. What we have is the inappropriate application of the metrics, the skewing of the measures and terrible results. The measures are skewed to give the appearance of success, but the reality is a Program in free-fall. Worse yet, the programs are becoming so completely focused on the near-term deliverables that the real tail-spin, is being pushed out in time. We will get to a future where we will have invested nothing and this lack of balance will doom us.

The first two posts show the real dichotomy existing at the Lab. Post 1 (7/27/2005 12:54:58 PM) is on the money in my view. While the program managers aren’t necessarily clueless, the system they are working in pushes them to act like they are. Post 2 (7/27/2005 03:35:25 PM) is the programmatic response “it's done every day in the Real World.” Of course most programs in the real world aren’t managing nuclear weapons! This is why its important to ask, the question about whether we are being effectively managed. For another take on the answer, look at the sort of briefings being given by Tarantino to his masters. See the Pad and Wipe website for an example. Its all metrics, nothing technical! I think its time for the chorus of “where’s the beef!” to be heard.

Now, the numbers associated with the Milestones. Here is the guidance I’ve seen: $10 million per Level 2 milestone, $3M per level 3, $1M per level 4 and $350K per level 5 (i.e. 1 FTE). I’ll just assume that the factor of three reduction per level proceeds for Levels 6 through 8. This works out to about $100K for Level 6 (4 months funding), $30K for Level 7 (1 to 1 ½ months funding) and $10K for Level 8 (2 weeks funding).

Let’s take a nice round figure of $1 billion to figure out what sort of numbers this gives. We will have 100 Level 2’s, 300 Level 3’s, 1000 Level 4’s, 3000 Level 5’s, 10,000 Level 6’s, 30,000 Level 7,s and 100,000 Level 8’s. Of course this begs the question what will anyone do with all this information?

Here is one theory: “if you can’t dazzle them with brilliance, baffle them with BS!”
6:35 is on the money when he said: "It has become so important to succeed with all the milestones that the program makes the decision to dumb-down the milestones to make them easy."

At DARHT axis II they were failing to meet so many milestones that their unofficial motto became: "If at first you don't succeed, redefine success."

They started out with the milestone of producing radiation, when that couldn't be realized they fell back to hitting the target with a single electron, when they couldn't do that they settled for just getting the hardware bolted to the floor and not plugging it in. Then they declared the project a resounding success. That's project management Los Alamos style. It is based on BS.

Here's the sad part. The axis II team worked nights and weekends to meet unrealistic goals and only succeeded in bolting a large non-functional machine to the floor. Immediately after success was declared in Washington they ripped it all out and took the proper steps to fix it after the political heat was off. At this time all indications are that axisII might turn into a success because of the team working the real problems. If they had not had to jump through the PM hoops they would be a year ahead of where they are now and the taxpayers would be $15M richer.
Modern project management methods work lots of places, just not LANL. And clearly they aren't working so well at LANL either. I think the methods work, but not the way they are implemented at LANL. The truth is that if a project is large and it is discovered that it was a bad idea from the beginning, no one can admit it because admitting the failure is a career killer for whoever admits it. Killing a funded program that is worthless, but brings in money is another career killer. And we have all seen one case where the people who discovered wholesale theft at LANL were fired.
The problem is that the lab is managed by cowards and turf warriors. Huge sums of money are wasted on projects that everyone knows won't work and on doing projects twice because no one wants to give up a funding source whether it is needed or not.
The real problem at LANL is that so many top managers aren't as interested in their project's success as in their own personal success.
We need management that cares about getting the job done. If we get that, project management will work.
To 3:03: Amen.
Acknowledgement that a major project is useless is career-limiting behavior.

Indeed, at most goverment laboratories and major govennment contractors, the measure of success is how much money the project brought in. Nothing else matters.
Here's the sad part. The axis II team worked nights and weekends to meet unrealistic goals and only succeeded in bolting a large non-functional machine to the floor.

Sounds like the NMT to the rescue-we know better than anyone-people are our most important resorce unless you tell the truth SST pad. Talk about productivity. We had our fifth emergency exercise today-don't know if DOE is happy yet. We built a two million$ carport and broke every rule in the book. And wonder why the DNSFB thinks we put program before safety. Productivity? Insanity?
Post a Comment

<< Home

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