Got it. Pretty interesting (& creative) approach.
We are still implementing (403D) also (not likely live for a few more months).
We have 2 distinct production areas:
A very short cycle time/unit Assemble to order area (which generates the product fulfilling customer orders) and is being migrated/reconfigured using lean principles.
A full machine shop the produces strategically best-produced-inhouse components (and acts as a backup to our world wide vendor sources).
The machine shop is a mix:
CNC equipment (setups 1-6 hrs - typically about 2 hrs, and lot runs from 3 pc to thousands -average OP taking about 3 days to complete the scheduled lot qty).
Manual equipment (drill heads, tappers, a few old bridgeports, deburring, semi-automated (press) assembly and stamping, and a paint process. These are all minimal set up very short cycle time OPs (with the exception of paint which has some inherent drying time).
Like you, we intend to backflush all our assembly labor to the multi OP standards - only requiring a last OP qty complete report via scanners and MES to handle the 175-200 jobs processed daily with minimal non value added reporting time overhead.
Our manual equipment shop areas all are planned to be qty reporting only (resulting in cost based upon the OP standards). (Manual MES initially but I suspect scanners will be added to improve entry acccuracy.)
CNCs, we intend to track by time & qty reporting (in the hope the traditional data will enable us to ID opportunities for process improvement via training and/or tooling/programming optimizations).
Unfortunately (from my perspective), we aren't going to be using MES for time/attendance as we have a 3rd party hand scanner/ADP process accounting is happy with (but will limit what we can glean from job activity reporting).
Your (?or 'management's?) idea of time/attendance use for indirect activities is pretty neat for a fast paced, cell environment.
I would think scanners could make it rock solid (better accuracy) and more likely to reflect 'real time' as a result of it being faster & easier (once all are trained to use it).
"Man minutes/part" is great if the cells aren't excessively mixed model flow (with potentially quite disparate total unit labor times produced in the cell - even though you may have balanced all operations for a steady output rate from the cell).
My desire (in our assembly area) is to have daily/weekly/monthly measures available on demand to measure sales $ value assembled to actual labor incurred over the selected period (as it is the only true profit center with value added activity). The decision to not use MES time/attendance complicates that goal tremendously (as it has also done our our legacy system) - but perhaps we can get our attendance scanners (and the db/apps they are driven by) to talk to vantage.
We have just started converting our manual machine shop areas to kanban cells versus the current 'traditional' OPs on complex mulit-op jobs. Once we reach a critical mass of change over success, I think your "man minutes/part" measure would be quite applicable to us.
Thanks for the additional info. It really helps to hear how others are (or intend to) use MES as it seems it is a rare subject (and epicor consultants tend to skim through it instead of offering ideas of how others have used it for real advantage to ID waste and improve producivity, increase effective capacity and throughput, etc.,). Your intended use to use MES time/attendance in conjunction with non-value added activities is really intriguing - as that IS the very fat in the processes that needs focus in order to reduce/eliminate it.
I'd really do some investigation on whether scanners would be a quick ROI for you. Properly set up, they can really enable individuals (and teams) to take more responsibility for their work. Giving them access to your process measures (perhaps suitably tweaked to match the audience more effectively) - can really lead to people/teams taking very active self responsibility for maintaining standards (and ecourage the sharing of their ideas for continuous improvement).
Rob Brown
brad_boes <
bboes@...> wrote:
For a very different perspective...
We are implementing 8.03.403 (soon to be .404) and do not intend to
use the job costing as such. We intend to use the MES time and
attendance with all of the time going to indirect activities that
are organized around shop areas.
Then the job selction (mostly cell time work) will be handled by
supervisors and/or lead people. The earned hour labor will be back
flushed and the actual time pooled in the indirect buckets. then we
intend to marry up the cell production verus the indirect bucket at
various points during each shift.
If you don't need the job by job costs, but more of a macro
productivity/cost look this could apply to you as well. We measure
productivity with a metric called "man minutes/unit" in each cell.
We do this now in our current non-epicor system and it works well in
a lean, cellular environment with lots of fast-paced, smaller jobs.
Hope it helps.
Brad Boes
Metalworks
--- In
vantage@yahoogroups.com, Robert Brown <robertb_versa@...>
wrote:
>
> Just what is it that 'management' is trying to accomplish by
insisting on MES being used like this? (Maybe MES isn't the best
solution to accomplish their goal.
>
> By having a few people do non-real time entry, you are creating
a process where the entry people can artificially make certain
people or processes LOOK better or worse than they actually are
(thwarting any attempt to use MES entered activity data for process
improvement).
>
> These entry biases can be unintentional (based upon a
supervisor/entry-person's assumption of someone else's typical
behavior) and intentional (creating potential legal issues if the
data was used to punish or reward individual performances).
>
> If you are using actual costing, the biases impact the G/L and
possibly inventory asset valuations subject to tax.
>
> If MES is the best path for fulfilling management's need, are
handhelds with scanners (which, if set up at appropriate work areas,
would allow real time, accurate entry) a viable option?
>
> Rob Brown
>
> Adam Long <xcsdm@...> wrote:
> I forgot to mention, we are running Vantage 8.03.403D
>
> --- In vantage@yahoogroups.com, "Adam Long" <xcsdm@> wrote:
> >
> > Per management decision, we are running MES outside of its
intended
> > purpose. MES is designed to be used by the shop employee "doing
> the
> > work". Our physical layout of our shop floor limits the number
of
> > computers available for employees. Since our jobs/work is
> scheduled
> > by the supervisors anyway, we decided that the supervisors are
> > responsible for signing their people into and out of jobs.
> >
> > This presents an issue for us. First thing in the morning, we
have
> > 60 people standing around waiting for work. We are looking for
> > suggestions to more accurately enter start times on jobs.
> >
> > 1. Is it possible to edit the start time of a job (From MES) ?
> > 2. Is it possible to add a field to the start production
activity
> > (and most likely all "start activity" dialog boxes) that would
be
> the
> > start time?
> > 3. Any ideas to get a more accurate time/cost per job?
> >
> > In many cases shop workers can be on a job for an hour before
being
> > logged into the job within Vantage/MES.
> >
> >
> > Any suggestions would be appreciated.
> > Adam
> >
>
>
>
>
>
>
> ---------------------------------
> Be a better friend, newshound, and know-it-all with Yahoo!
Mobile. Try it now.
>
> [Non-text portions of this message have been removed]
>
---------------------------------
Be a better friend, newshound, and know-it-all with Yahoo! Mobile. Try it now.
[Non-text portions of this message have been removed]