Notes from the Deployment Front
There are just three things leaders need to get right.
First is to act like leaders. That is to say when you are involved in this be clear about your commitment. Any leader that talks about this and then talks about more important priorities is not a leader. People hear actions clearer than they hear words.
Second is to align projects with disruptions in the key processes that are aligned with strategy. Simple – look to Policy Deployment for guidance.
Third is measure the right things, but first get your measures right. This is different than MSA and GR&R. Other than financials, you want to measure quality, time, and cost of each key process. I will talk measures of time for the next several posts, because I believe that fixing responsiveness also, by necessity, improves quality and cost.
Supplier Responsiveness –
I believe that any enterprise that depends on supplied resource (goods or services) would agree that the ability of those outsiders to give us what we want; when we need it is critical.
Most companies have two measures – on time to request and on time to agreement. I think on time to request, assuming lead times are defined and respected in the request, is the only important metric. I’ll argue that another day. Regardless of which you drive to or if you drive to both, it’s what is behind the numbers that makes all the difference. We can choose to measure to make people feel good about a bad system or we can put a straight system with simple rules in place that will initially give us ugly numbers. It’s our choice and it really goes to leadership – are they really committed and improving or complying with some initiative brought in by the CFO that just joined you from GE?
A simple example –
Say you work for a multinational company (they usually call themselves global but still make all decisions from HQ) and you have been chasing less expensive labor and materials to Latin America and Asia for the last five years. Odds are that the old purchasing metric of recvar looks great but you now have long lead times. Odds are that your forecast horizon is inside of lead time. Your purchasing and planning has turned out to be the expediting department. Your factories are carrying lots of inventory and your delivery to customers is poor. Sound familiar? If not, don’t read further.
Take the order you just sent to your new friends in Bangkok. A twelve week lead time and you realize you need it in six. Transport time is four weeks and so you change your request, completely disrupting their factory (for the fourth time this month). They come back and tell you that instead of the 100 units you wanted twelve weeks from now, they think they can give you 20 every other week starting in 6 weeks. That means that instead of 100 in twelve weeks you get 20 six weeks early, 20 four weeks early, 20 two weeks early, 20 on time, and 20 late. What is the measure of on time performance?
a)80% – we got 4 out of 5 deliveries by the original request date.
b)80% – we go 80% of the material by the due date.
c)20% – 1 of 5 deliveries was within our on time window of +2, -2 (not knowing what is coming daily makes it really hard to clear receiving every day).
d)0% – we needed 100 and not having all 100 really disrupted our internal operations.
e)100% – they are trying really hard and it’s really not their fault.
The only answer that will cause us to fix our system is d). What do we need to fix? Go through the five whys and we have to get back to you cannot plan inside of lead time. Either get a better lead time or extend your planning horizon outside of lead time and then live with it. Anything else causes churn inside your extended enterprise and putting a feel good metric out there will not help a thing.
The one thing I always tell people about Six Sigma is that following the method will guarantee improvement with two caveats – the project work has to be supported (time and team) and you cannot violate the laws of physics.
Planning inside of lead time is violating the laws of physics, lead time tells you your planning is already finished. Creative metrics will not fix the system!