Hot, cutting-edge companies

January 19th, 2009

Acme Tool Co. still calls layoffs “reorganizations.” How lame. See how cool companies do it–compare two Google announcements.

Pay attention to the titles–this is CHANGE, people! Oh boy!

First engineers get changed, then People Operators (see signature) get changed. Shortchanged?

Of the two, I think I prefer the changes to the People Operators. It basically says, you’re fired, it sucks, sorry. The one for the engineers trys to explain it away and it scares me pretty bad when it says “Having offices distributed around the globe […] presented unique challenges. The most difficult of these being to […] provide engineers with significant, meaningful projects.” Meaning? We hired you and we didn’t even have any work for you to do! Man, I missed that gravy boat.

Pity Party: No Cover Charge

January 7th, 2009

Welcome to the Pity Party. It’s easy to get in; no cover charge, just bring a complaint.

Problems we have legion. About a year ago the A-Team left, having brought with them a Mandate from Heaven which we were Not Allowed to Question and having left behind, in several incompatible versions, a process which added more time, touchpoints, and inventory to our build process. About four months later, in the throes of learning focus all of our efforts on meeting our first promise to the customer, we were given a stringent demand to improve in three months or face dire consequences. For months we learned, bit by bit, how each seemingly inconsequential part of our process affected the other parts. The total sum of our software idiosyncrasies gradually distilled into the only internally cohesive process we could fashion, and, as our deadline drew to a close and we met the goals established for us, senior leadership fired our plant manager. Then, after another month or two, they fired the man who had driven the plant manager to undertake all the painful changes.

Then they announced that we would go back to striving to meet the customer’s request rather than our first promise, as we had done two or three years ago before they began the ever-intensifiying push to meet our first promise. Most recently, it was announced that we would change our build process to resemble what it had been before the A-Team came to enlighten us all. Meanwhile the existing process is failing because they laid off the people who had been proficient in it and retained people who had not done it before.

Bookings have collapsed to about half of what they were a year ago but past-due orders have been decreasingly only fitfully, held up by unexpected sales of product we can’t build due to breakdown of equipment that we cannot replace or effectively repair. Other tools could be built but we do not have the capacity on the assembly line (which was two assembly lines before being efficiently combined by the A-Team) or enough people to bring the material to the line–at least people who can do so quickly, since those who were familiar with this line were laid off. Other tools can’t be built because, although planners have been admonished to make sure all material is available five days in advance of when it is needed, the new ship to customer request policy means that while you are holding four components in anticipation of a build (and possibly waiting for a fifth), any or all of the four could ship as spares to orders that come in requested the same day. Then you must call your supplier in Japan or England or China or Russia or India and inform them that your schedule from two months ago turned out to be incorrect, and you need some more parts no later than four days ago.

Meanwhile, we are probably going to switch suppliers for crucial raw materials back to a supplier we had years ago, whom everyone who was around then desperately hates due to their abysmal service at that time. But the people who make decisions now were not around then, and the supplier swears they will deliver on time, and at tremendous savings over their competitor. Another supplier went out of business and closed their doors with no prior warning, leaving us scrambling to find a replacement. But all is not bleak on the supply side: after years and years of poor service from one supplier, we are finally going to switch–if we can convince this notorious ne’er-do-well to give us back our tooling used to make our parts without paying exhorbitant service and maintenance fees.

This is how today went: First thing this morning, our Process Improvement expert (who was our key ambassador to the A-Team, and the one who reoriented the assembly lines how they were before, and just announced that we will use the process we were using before, and is the key person in giving all the reasons and data for how we will change things to meet our goals) called me over to ask question on how our inventory allocates to our sales orders. It turns out he did not understand how that process worked and had therefore used wrong assumptions to calculate safety stock levels. Consequently the plant has been doing an abysmal job attaining to the safety stock levels it has committed to. After that I went to a meeting that was not really a meeting; it was the evicserated remains of the meeting we used to have whereat the planners got together with the work leaders to finalize the plans for the day. Here unfortunately the planning manager expressed his general dissatisfaction with and lack of understanding of the decisions of his supervisors, with the resultant lack of understanding of what they were supposed to do and how they were supposed to achieve it. It was the type of meeting where you try to be invisible, but before it adjourned I had to inform the planning manager that one of the tasks for his group that I had tried to minimize I was told had to be done in an extensive fashion–although really I am not sure how it is supposed to be done at all, since a variety of hazy and contradictory messages have gone floating through the management level above me, leaving me uncertain if the task can even be done with the tool I have spent the last four months gradually refinining for this purpose.

Then I went to a second meeting, the reincarnation of the meeting we used to have where the plant manager asks his management team questions they are ill prepared to answer, they having been accustomed to the presence of their subordinates who know the details of the situations at this meeting. Here, the failure to build some tools was blamed on the lack of some parts; the person responsible for the constraint process said it was an exceptional failure lasting less than eight hours which would soon be resolved. The plant manager asked him why the parts were not ready five days in advance, as he has been insisting upon; and he has indeed repeated this frequently and clearly. This particular group of parts are reasonable easy and reliable to make, fairly expensive, and used in large quantities; the process manager had delivered exceptional inventory turnover (one of the metrics that had been stridently promoted in the recent past, and an overall cost saver) by running a just-in-time process that actually used a Kanban, or replenish when used, system, unlike other so-called Kanban “pull” systems that exist more in trendiness than reality.

The planning manger, still sour from earlier in the morning, obdurately pointed out that the safety stock number in the system was not maintained, and it was a mystery to him why. The manufacturing manager, who should have stood up for his subordinate, sat quiet while the plant manager cross-examined him, but responded forcefully to the planning manager’s accusatory implication, and was heard later in the day vowing “no more Mr. Nice Guy.” Without support, the person who had maintained an actual demand-driven, high-velocity process conceded and pledged to produce mountains of inventory so that we may never run out again.

A short while later, I caught the tail end of a meeting about how to maintain our safety stock levels under the influence of our allocation process, a follow-up to the conversation that started the day. The options on the table seemed to be either 1), build mountains of everything so that we can allocate a week’s worth of orders and still allocate any sudden demands that come in for same-day shipment, 2) let inventory that was built for a customer who ordered in advanced be shipped to a customer who ordered same-day, or 3) get someone to constantly intervene in the automatic allocation to game the results to what we wanted. One person mentioned that it seemed like we should not use inventory to hide process problems–the same planning manager who had carelessly accused the work leader of keeping inadequate safety stock.

The rank-and-file planner who was brought into the meeting left clearly downcast. I did was not present for most of the meeting; I don’t know what had already been discussed. I can’t imagine a way he could have left without feeling like all the layers of management above him had no idea what they were doing. For months we had trained our previous plant manager on how things currently worked and how they could be made to work with our current systems–and he, in turn, and passed along some education up the ladder to the man who ultimately fired him. Now both were gone, and we had a new pair that needed to be educated–and none of the people who remain in the layers in between seem to remember anything from the last time around. Once more we are learning how a three-legged stool works by shortening or lengthening one leg at a time.

There is one thing that everyone understands–plant manager, staff manager, department manager, salaried employee (if I ever ran into hourly employees any more I am sure they would also)–and even the new Vice Admiral, the new mover and shaker. They all understand that our systems way of assuming that a customer is requesting an item the same day if they do not supply a date is unfair. Managers and plant managers have told me that it makes no sense. The new manager in the position that I used to report to directly began to get a little loud as he told me that people’s livelihoods were on the line and it was no matter to take lightly. (Do not conclude I am an important person by the people I converse with — I represent a manager based elsewhere and sometimes get the initial reaction to policies or decisions that are not popular, but I do not control any policies that affect any of these people.) If we assume that customers want everything the next day when they don’t specify, and the majority do not specify, and we are being measured on how well we meet their (presumed) request, we are setting ourselves up to fail!

I agree that it would be better to default to our stated lead time, giving the customer the option to override if they have some other wish. But there are much bigger problems to be had. If every manufacturing facility is being held to the same standard (and they are), it is unlikely that they will fire us all and close all our plants if we all attain roughly the same success by this measurement. If we work to demonstrate, explain, and document why we can’t meet this level of service–what long lead times we have by using overseas suppliers, what unpredictable response times we have with aging machinery–we can build the evidence that might convince distant high levels of management to help us do better. But we can’t treat our inability to achieve this high goal with the finality of death. The “It’s impossible” attitude was the seed of death for American car manufacturers, and it will be ours, too, eventually.

The last thing we need–and the first thing we seem inclinded to do–is to attempt to meet this impossible goal of shipping anything the same day by building mountains of inventory. You always lose with this strategy. A model becomes obsolete, you discover that some component was made wrong two batches ago, or you just don’t build quite enough of some unexpectedly popular item–you still fail. This calls for Lean manufacturing–and that does not mean firing everyone! It means cross-trained workers who can do different jobs depending on which jobs most need doing at this moment. It means machinery that is multifunctional or is rapidly change over, so that if customers actually want part B instead of part A we can accomodate. It means investing time and money into understanding what makes your machine break and what can prevent it from breaking. It means accepting higher costs for suppliers who can prove and maintain their delivery so that you can retain customers with solid promises, not attract them with lofty promises and lose them with repeated let-downs. It means a miminum of inventory in a minumum of places that requires a minimal amount of moving around, counting, touching, handling, and passing off to fashion it into the product the customer ordered. And we have the perfect opportunity to show how it is impossible to deliver a complete product in one day that relies on a component that takes 90 days to get from Pakistan or Israel or Korea.

But we dare not let ourselves do any worse than we must at the utmost limits of our imagination, and it seems that if we just had a lot of everything we might be able to ship anything at any time. But it is so unfair that they make it all the harder by assuming that customers would like to have everything the same day!

Getting in the pity party is cheap. Getting out is what costs you your job, your business, your industry.