Logging production nfrom 80 wokers in a unique type of flour mill

mickal

Currenntly there are mistakes as all the woekers write downn their ibdudivual packing of serveral products and that then ngoes tom the office to be input into the ncomputers. Mistakes are made and total productiom usually mhas errors.

I was thimking of simplifying this process by using smart cards or rf. The bworkers would possibly have keypads to punch in the basg sizes 500grams to 40kg/ Thenn theuy punch in the number of bags theyb actually filled and it could nbe sent directly to a computer. Thereby the inventory would be more accurate and alsoi the production of each employeee. Its not going to be used to check on workers, more improve inventory recordes.

Any ideas here would be appreciated.
http://www.kiallafoods.com.au/About/about.aspx

Mike.

ctwardell

Mickal,

Sounds like a fun project.

We had a key entry system at a major shipping company where I worked in the late 80's to early 90's.

The purpose was to capture destination and quantity of small packages placed into "bypass bags" that where used to prevent repeated handling of the indiviadual packages in intermediate locations along the shipping route.

The system worked fairly well, but workers had a tendency to inflate quantity numbers to make it look like they had higher production.

I have a few questions:

Is the existing packaging specific to the product it contains and if so does it have a barcode?

If the packaging is not product specific how do they indicate what it contains, a tag maybe?

If they use tags could those be barcoded?

C.W.

bill190

Actually I would look at designing an entire system. Perhaps an order could be kept track of with barcodes or labels printed to place on boxes?

Maybe electronic scales could be used to do some "inputting"?

Looking at a video of what the workers actually do would be helpful.

And the layout of what is done where would be helpful.

Basically you get an order...
[what all happens?]
...then in the end there is a box or boxes waiting to be shipped.

FYI - You might want to read a book about Henry Ford if you are wanting the workers to be more productive. He was a student of "wasted motion" and figured out that his employees were spending a lot of their time walking around. Thus the assembly line was born and that greatly reduced the walking around. (But it also created employee burnout!)

Also if you are designing the system, you should do this work yourself for a day. That can give you a lot of insight into what the system needs. And ask the workers what they like / don't like about the current system / proposed systems. That is the most important person to ask - they guy who does this work 8 hours a day!

mickal

1.Is the existing packaging specific to the product it contains and if so does it have a barcode?

There are hunderess of different products.I worked there 10 years ago part time and setup the barcode printing machine which prints cardboard barcodes and is sown onto 20kg andn 40 kg sacks of all kinds of differnt products. They didnt scan the barcodes ,think they still dont. They used the number of printed barcodes of each type, counted the number of barcodes that had been printed and used that to for inventory levels. That was alittle error prone due to the barcode machine having printing problems.

Smaller items of around 250 grams or 500grams 1kg , 5kg were just put in boxes and the box was sealed and a barcode was put on the box. But no one new how much of the smaller items , without much accuracy untill they were boxed up.. If the packaging is not product specific how do they indicate what it contains, a tag maybe?

I dont even think when large orders are made up, despite many bags having barcoded tags they they are ever scaned. Like I said the tags are used by counting how many of each kind of tags are printed and that tgells them roughly the stock. Its not a great system.

They do have problems now and then of their customers not getting correct orders.

3. If they use tags could those be barcoded?
Perhaps they could be talked into getting barcode scanners escpecially for the men loading up all the pallets with product and scanning the barcodes to ensure the order was 100% correct.

4. I still think that that say when workers were say filling 25kg bags they could set the keypad to 25kg and every time they filled a hundred bags they could punch in 100 on the keypad. Then if the machine that printed the tags was playting up they would know to trust the operator and his input first.?

5. The place is bigger than it looks. There is room for a better way. This is just a very rough outline as to whats going on. I made the website :-) They get orders all over the world.

Can you expand on my outline or come up with other ideas v?

mickal

Its not just inventory error problems, its also the workers making mistakes as to what thety are loading onto pallets for shipping out.Iwent with the owners to see abig oberation in a different industyry. They had nil customer order errors by using barcodes in a far more efficient manner. Also in truth a keypad could be used to see how many of each product type could be produced and packaged in a day buy each differnt bagging station.

I might endup going back with a video camera :-)

localroger

mickal, designing systems like the one you need is my job. We primarily focus on systems that use weighing scales because that's the hardware we sell, but I have done this kind of integration with the scanning, key entry, and report generation for many different applications.

I usually insist on walking the site before even making a quote on this kind of job. It's still not totally clear to me exactly what you're trying to do here, and my experience is that it is the edge cases -- the little exceptions that nobody remembers when writing a specification -- that make these things hard to do right. I like to watch the operators work, watch what they're currently doing, then ask the actual guys filling out the slips what they do when this or that might happen.

But before I would even get to doing that, I'd ask a few other questions on which I'm still not clear, and which greatly affect the difficulty of the job.

1. You need to list all of the data inputs to the system and where they come from. What information do you want to record about each bag, and how does each item get entered? Do you scan it? Is it key entry? Is it setup beforehand for a run of identical product by a supervisor? Does an item change infrequently so it can be set and defaulted, or does the operator need to be prompted for every bag?

2. Then list all of the data outputs -- what reports you want, what statistics you want to keep, and where you want this data to be available and how it's going to get there. Do you just want to print it out on the production floor (printers, ugh, but no communications)? Serve it up as a webpage so it can be printed in an office? Just run an application on a PC that runs everything and prints the stats, but only from that box?

3. What kind of user interface do the entry terminals need? Will the operators have a problem reading prompts and operating a user interface that consists of more than just a big green button?

4. Can the terminals be fixed in place or do they need to be mobile? The latter is much more expensive, complicated, and prone to theft and loss.

5. Is there a commercial product suitable for being a terminal, or do you envision building something? Since we work with scales we have several programmable instruments we tend to use for this kind of field user interface. Unforunately these are high-end instruments costing $1000-$2000 each, but they are self-contained and ready to go. And they're legal for trade scales. (I am currently working toward making a cheaper Propeller based terminal for us to use on this kind of job, but it's not likely to be ready for a year or more.)

6. Is there wiring in place to connect the terminals to a server of some sort (which might not be a PC style "server" but just something that collects and collates all the data)? If so, what kind, and if not, what kind would you prefer to run? We use RS232 for almost everything of this nature, but some people really prefer ethernet and sometimes the RS232 needs to be converted to a more robust level like 20mA current loop for optical isolation and signal integrity on long runs.

7. What do you want the system to do when there is a service interruption (cut cable, failed terminal, server down)? Barcode scan fails? An incorrect code is entered? An incorrect but valid code is entered and bags are run and then the mistake is realized? If you want to be able to edit the data to correct such errors where do you want the user interface for that to be, and does it need to be protected by a password or hardware interlock?

Based on what you've said so far I would guess that my company would quote you in the range of $5,000 to $10,000 for programming services (basically my time) to create, install, and debug a system like this, plus the cost of hardware. That reflects a week or two of my time. And no, I don't make that much myself; we're an industrial service company and there is a lot of expensive infrastructure we have to cover. I've shown up to jobs in the passenger seat of a $250,000 test truck that exists solely to drop calibrated weights on truck scales. And I've spent a lot of unbilled time over the last 25 years doing R&D and educating myself so that I can do a job like this in only a couple of weeks and get it right with a minimum of fuss.

If you want to roll your own it's a good way to learn but if you've never done anything like this before it will probably take several months and some false starts to get going; I'll be glad to give you as much advice as you can stand.

mickal

Thanks LocalRoger, that was a lot of 'white paper' to get my head around. I was a very small parter in the business , hence I called the other two fellows my bosses:) He gI ended up leaving and it njustn grew andn grew.It started out with one relative we thought was a little, how shall I mput n. He seemed to march to a differnt drum(i think i do do) He started in a tiny shed with an old mstone wheat ngrinder ans put a very small mhonda motor on it to turn the grinding blocks and make floure.About 100kg a daym he managed. Then nhe kept adding nthings, biger grinder andnmotor, extended the shed and it just kept getting bigger and nreally cramped. He was a sucefully farm, a very unique man, would not allow sprays or artificial fertilzers etc, That actually worked, after a fewm years he was getting much better crops than his neibours.
Then he took a big gamble and loaned money and built a huge flower mill and hes never looked back.

I am busy for a month or two but I will have a lot of answers when I get back from mthere nto post.

Thanks for yourn information.