production planning & control for fabrication business

[JosephERG]

Good Day!!!

I would like to seek advise about the best production planning and control system for a fabrication type of production. Currently, our company is using the "process yield" as basis for production planning. I think that this approach is more applicable for mass production type of business. The yield per process that can be generated from mass production is more accurate because the quantity is in volume and the product types and its processes are generally fixed.

The fabrication type of business is the exact opposite. The quantity is very small (in our case, orders are usually less than 10 pcs), the products are very varied and so are the processes. The yield per process may not be representative of the actual process capability.

This is the reason why, despite setting a cut-off yield per process, we still fall short of our production targets. Most of the time we experience delays in our production schedule.

Any advise you can give? Thank you very much.

 

[tripleZ]

If I'm reading this right, you're basically a job shop that is having trouble meeting ship dates? If that's the case, I'll be looking to see what folks suggest.

Having run a tool & die shop in years past, production management was difficult to do with conventional software packages. Why? Because EVERY job was unique, and often the manufacturing process could vary from the quote. You have a set bank of machines that can be used for a variety of functions. Thus your bottlenecks are inconsistent and vary on a daily basis. If you're a lean outfit, your operators likely perform multiple tasks, meaning downtime can occur at any time. And the time it takes you to do all of the data entry in a canned software program can often take more time than it's worth, especially if you're constantly updating the thing real time. The list goes on.

How big is the company out of curiousity? Sales?

 

[tomwalz]

That's the way it works here. Our big problem is suppliers. After a while we learned who we can trust and by how much. We build as much cushion for time and expense as we can. We multi - source whenever possible.

We never hit yield targets.

Many years ago I was doing some work with a non-profit. Whenever they were asked to bid on a new job they set up a work station and the boss sat down and did five parts. (Sometimes he would work for three minutes or five minutes.) Anyway, that was the standard. The staff never came close to meeting it.


Thomas J. Walz
Carbide Processors, Inc.

http://www.carbideprocessors.com

Good engineering starts with a Grainger Catalog.

 

[JosephERG]

Thank you guys for the valuable insights. Its good to know that we are not the only one experiencing this difficulty.

Our company is engaged in the manufacture of precision tools used for semi-conductor processes (i.e. die bonding, wire bonding, etc.). Our customers are mostly from Japan and the US.

I am the engineer in charge of manufacturing processes and systems improvements, so I do not have access to sales information.

By the way, how were you able to work out your production planning & control system despite the obvious difficult nature of the fabrication business? Care to share it with me?

Thank you very much.

 

[PSE]

JosephERG

You will likely end up "customizing" your equipment to suit each customer's need at the time. These may change within the same base customer depending upon the fab where the equipment will be used.

For your consideration I would suggest the following. In each of your tool designs there will be common components used in its fabrication. These will have a lower probability of change based on an individual customer's requirements and may be suitable for production forecast type of planning (X number of systems per time period with Y number of systems in the same period = Z components needed for build purposes).

Then you will have the unique components/configuration for each customer. I have used successfully in the past a tracking file with a customer code and a unique tracking number (using the date the project starts works well). Here you can capture and record requirements/costs/times for the customized portion of the product. I always kept the "original" file and sent a copy of the file along with the system as it was being built. It allowed capturing the unique information that might be useful for followup and service/maintenance events. At the end, I maintained the master and reference file for each tool. Subsequent orders started to paint an overall picture of the customer and it became easier to anticipate what they would be expecting.

Good luck with your efforts.

Regards,

 

[tomwalz]

We build custom tools as well. In our case it involves brazing carbide or ceramic to something. Each job is different. Part of what we sell is our existing knowledge and part is our ability to acquire or develop new knowledge. However no one wants to pay for that.

We take the amount of fumbling and screwing around we know will be required and build it into the price. This is pretty well a guess. After enough years we are pretty good at it.

Manufacturing never hits the ideal. Short run is even harder. And custom short run is harder than both of those.

If you haven’t done so I would suggest you read The Goal by Goldratt, Made in Japan by Akio Morita and several works on and about Deming.

Tom


Thomas J. Walz
Carbide Processors, Inc.

http://www.carbideprocessors.com

Good engineering starts with a Grainger Catalog.

 

[ShopPlanner]

JosephERG,

You may visit

http://www.optisol.biz

if you are facing any difficulty with scheduling of your custom manufacturing.