[chrisryn]

I want to know in your opinion. At the end of the day what is more important part count or machine utilization in a production shop. After I get a few response I'll give the results of a test I conducted.

[RICHARD ZASTROW]

Why not both? Are they mutually exclusive?

Dick Z

[Geof]

How can you separate the two?

Imagine that at the end of a day you have a part count of X and a machine utilization of 50%; obviously if you improved your machine utilization to 75% your part count would go to 1.5X.

[ImanCarrot]

Bit of a daft question if you don't mind me saying.

Part Count dun matter if you're not making X profit on each part, machine utilisation is only important if it's making you money- spend 10 hours a day making stuff that isn't making you profit and you'll know what I mean. Factor in electricity, wages etc etc, you can have a machine running at a loss all day and that would not be very good, would it?

If you make 1000 parts a day at 1 pence loss per part then that ain't too good. make 1 part at 1000 pence profit and well, I think I make my point.

The both are tied together.

[chrisryn]

Here is why I posted this question.

I did a simulation run of 1 hour

Run 1 was orginal program, fixture, and procedure.
It resulted in 80 parts ran in an hour with a 50% utilization.
Cycle time was 1:34. Load/Unload time was 1:32

Run 2 was new program. Fixture and procedure was same as run 1
It resulted in 120 parts ran in an hour with 20% utilization.
Cycle time was 33 secs. Load/Unload was same as run 1


So in a production facility which run is better.
I say run 2 since it put more parts out the door.
Others say run 1 since all they care about is utilization.

Thats my argument, and why I posted this poll.

Edit due to typo.

[Geof]

I concede ; improving part count with a lower machine utilization seems counterintuitive.

[g-codeguy]

You are looking at it by comparing only how long it takes to run one set up. How about using the number of set-ups run in eight hours for a comparison? Looked at that way scenario number one could theoretically run 160 set ups while scenario number two could run 230 set-ups.

I would consider 2 to be a better utilization of the machine. How about you?

[ImanCarrot]

So in a production facility which run is better.

Neither unless you know whether you're making a Profit or Loss on every part that's spat out of the machine.

ie: if you're making a profit on each part then case 2 looks better, if you're making a loss then case 1 is better, unless you have employees that you need to keep busy at a loss so you got them for future work that'll make you money.

In actual fact... even if you're making only a small profit it's still not worth making stuff, you'll wear your machine out unless you've amortized the capital costs.

There's too many factors, but at the end of the day it's MONEY that matters

[chrisryn]

I agree in a typical shop that would be true. 99% of our production runs are profitable very rarely do we machine something and loose money on it. If we do its due to an operator failure and have to rework large orders due to scrap. So in this issue money isn't a factor.

[Geof]

Neither unless you know whether you're making a Profit or Loss on every part that's spat out of the machine.....

....you'll wear your machine out unless you've amortized the capital costs....

.....at the end of the day it's MONEY that matters

But read the detail of the analysis:

For Run 2 the machine utilization was reduced while the part count was improved.

Assuming the parts are making profit this reduces wear and tear on the machine and maximizes money at the end of the day.

Anyone who thinks that Run 1 is better is, to put it politely, not a good businessperson.

EDIT:

The only way to improve on Run 2 is figure out how to load the machine while it is running; i.e. use some type of pallet system so the changeover from a finished batch of parts to the next batch is a few seconds; then both part count and machine utilization improve.

[chrisryn]

But read the detail of the analysis:

For Run 2 the machine utilization was reduced while the part count was improved.

Assuming the parts are making profit this reduces wear and tear on the machine and maximizes money at the end of the day.

Anyone who thinks that Run 1 is better is, to put it politely, not a good businessperson.

EDIT:

The only way to improve on Run 2 is figure out how to load the machine while it is running; i.e. use some type of pallet system so the changeover from a finished batch of parts to the next batch is a few seconds; then both part count and machine utilization improve.

In most cases that is correct, but you have the pallet change that increases cycle time and will reduce part count in the end. This issue deals with a company that doesn't see the machining side of the company as a "Machine shop." The consider it a production shop. It is run completly different than most machine shops I have worked in. I would go into detail about the differences but instead I'll list the only simularity to this shop compared to a typical machine shop. They both have machines.

[Geof]

In most cases that is correct, but you have the pallet change that increases cycle time and will reduce part count in the end.......

Your logic baffles me here, and I am not trying to be argumentative, planning for best machine utilization is what has been keeping me busy for the past few years, I am interested in how you support your logic.

Yes there is the time to change the pallet but if that time is less than it takes to reload a stationary machine you will improve both part count and machine utilization. Your example went from 1:34 run, 1:32 reload to 0:33 run, 1:32 reload; if you could somehow reduce the 1:32 reload time you will make more parts. And even if the run time increased slightly, so long as this was less than the reduction in the reload time you would still make more parts.

[chrisryn]

Sorry I didn't put my reasoning for that statement. The part and fixture are to large to run on horizontal so they have to stay on vmc. Haas has a VMC with a dual APC. Pallet change time 30 seconds * 2 is a Minute so our cycle time is back up to 1:34 so your gaining machine utilization but decrease part count. Thats what I meant. Again sorry for the confusion.

[ImanCarrot]

About the profit margin thingy.

We occasionally make stuff at a loss whether it be: to get a foot in the door; or to snag the technical and proprietary rights to a job; or to keep employees that are key to our core business busy and not bored (the devil makes work for idle hands and all that).

A lot of our work is "hmmm.. will this work?" stuff.

I design the stuff I do, which gives my company the intellectual rights to that design. The person(s) who want us to make the stuff either have to get it re-designed or consider copyright infraction if they dun go with us after the initial loss- leader.. it's a risk, but it does pay off.

No offence meant!

[chrisryn]

I understand that logic and prefer that kind of work. The company I work for now doesn't do outside jobs the have products they develop and control from raw/cast to final assembly. So all they care about is how much went out the door that day. There spending alot of time chasing machine utilization when part count is there final goal. It's been an on going argument. Not to mention there are maybe about 3 people that have experince outside the company when it comes to machining. Every one else learned what they know here. Its us 3 against THEM.

[Geof]

Sorry I didn't put my reasoning for that statement. The part and fixture are to large to run on horizontal so they have to stay on vmc....

Ah, now I see we are in different ballparks.

1:34 must be 1hour 34 minutes.

I read it as 1 minute, 34 seconds.

[chrisryn]

Ah, now I see we are in different ballparks.

1:34 must be 1hour 34 minutes.

I read it as 1 minute, 34 seconds.

Your correct its 1 minute 34 seconds.

[Geof]

Your correct its 1 minute 34 seconds.

If I can continue to waste your time, this is intriguing.

Too big for a horizontal but run and reload time in minutes? Are you doing one part per load or numerous parts?

The description you give about the company you are with could be applied to my company; we design, manufacture and distribute our own products. For years I have been working on improving fixturing and programming to get more parts per day. I think you are on the right side of the argument, parts per day is the correct goal.

The way I have gone about this is to build fixtures that hold multiple parts: When the part needs operations on only one side the fixture clamps onto a fixed baseplate in the machine; kinda like a manually operated mini-pallet system, we have two fixtures and one is loaded while the other is running. When the part needs operations on several sides the fixture clamps onto a baseplate mounted to a rotary table so it can be indexed to different orientations. For some parts now our reload time for a fixture holding up to eight parts is shorter than the previous reload time for a single part when the fixture stayed in the machine and was reloaded with the machine idle. Not only that but tool change time is divided over multiple parts and for parts that previously needed three loading operations, with the machine idle, for work on three side now only need one loading operation.

Fortunately I don't have the US versus THEM problem; my background is in production machining, so I am on the US side and the THEMS in the office do things my way because I am the boss.

[chrisryn]

To awnser your first question yes there are multiple parts. $ on one fixture 5 on the other. The fixtures take up the entire table of our VMC. Now to take this even further I'll go into more detail. The machine utilization argument is on all of our machines not just one. The specific parts that I have been discussing are a trial I tried. The sad part about it is I now run 50 parts in the time it took them to run 10 a year ago. Nothing changed but program and tool. When I first set out to accomplish this I was told it couldn't be done.

Not to toot my own horn or anything, but I have been a machinst for a little over four years a program for 1.5 years. I am real good at cycle time reduction and setup time reduction. Solving these types of problems comes natural to me, but do to the time frame I have been doing this I get passed over alot on ideas I have even though I am the only true experinced one in the cnc shop.

Ok back to the parts once there run on the cnc they move over to a series of drill press were it takes close to 2 hours to finish the part to be ready for assembly. I had my tool and die maker make me a fixture that would allow for all secondary operations to be done in the machine but one(would require 5 axis). In order to make that fixutre stable enough for the secondary ops it can only hold one at a time instead of the 4 to 5. My production manager seems to think that our current procedure would be quicker than running all ops on cnc one at a time. He doesn't understand that more parts would be out the door quicker and utilization would increase do to longer cycles. Nor does he understand that the faster I make the cycle without change unload/unload time part count will go up but utilization would dramaticly decrease. Everyday it seems like I will loose this debate in the end.


Sorry for the long post.

[Geof]

No don't feel sorry, now I understand what you are up against; I feel sorry for you having to deal with people who are looking at the wrong numbers. We had analogous situations to your multiple drill press operations actually being more efficiently done on a VMC; actually for one part that comes of the lathe with a little tip from the parting the most efficient way to remove it is using a multi part fixture in a VMC. All that the VMC is doing is spot drilling the end to remove the tip and leave a litle dimple. I have had visitors think I am crazy tieing up a $50,000 machine doing something this simple, but the end result on the bottom line is positive.

Do you also run into the arguments about not holding too much inventory? Business schools teach the type of thinking that locking money up in excess inventory is wasteful; in some situations that is b.s. We get custom aluminum extrusions in mill run quantities to get a good price. We process three or four months worth of sales at a time, sometimes more, so there is often huge amounts of stock sitting in inventory. Again I have had people I know in business lecture me on my wasteful inventory practices; they simply cannot understand that the savings in material cost and production time far outweigh any loss of interest income that could be earned by the money tied up in the inventory. Mind you we can only do this because our designs are static; we know we will sell everything eventually.