JIT Production

While we are a job-shop, we have several customers that want us to
deliver parts in a JIT (Just In Time) manor. While we could plan
ahead (he says laughingly) and manage to have enough inventory
on our shelf to meet our customer's needs, in the real world (or, at
least, *our* world), this just doesn't happen.

We have several horizontal machining centers with multiple pallets,
and we've set up specific pallets for specific parts. We can then
program the machine to only run the pallets we need, depending on
the demand for the specific parts.

Vantage assumes that one sets up a job, runs it complete, finishes
it, then move on to the next job. When we have multiple jobs, all
running at the same time on the same machine (workcenter), and
the demand for those jobs fluctuate daily, how can we track our
time, costs, etc. for these jobs?

(Don't get me started on how Vantage divides the labor time equally
between multiple jobs, but applys all the burden time to all the jobs!
Plus, how we have to have our supervisors adjust daily <per shift!>,
what percentage of time is allocated per job <based on the specific
jobs run during that shift>)

How would one go about scheduling and forecasting? Vantage
assumes that the entire job quantity will be completed on a single
date, whereas we often have parts dribble in over a period of time.
If we set the job 'due date' to the first date we need parts, it's always
late, and we're constantly re-scheduling it. If we set the date to the
last date we need parts, Vantage shows that we're always below
quantities to satisfy our customer's needs.

I'm sure other shops have simular situations. How do you all handle
situations like this?

Thanks in advance.

Malcolm Spann
Malmberg Engineering
(925) 606-6500 x19
(925) 606-6505 FAX