Home > Forums
Topic Title: How to apply lean to aluminum manufacturing
Topic Summary:
Created On: 02/09/2007 04:10 AM
Linear : Threading
Send to a Friend Send to a Friend
Search Topic Search Topic
Topic Tools Topic Tools
View similar topics View similar topics
View topic in raw text format. Print this topic.
02/09/2007 11:15 AM
Print this message

Author Icon
128947
Reynold Murphy



I work at an aluminum profile extrusion plant. I have been asked to start a lean initiative. The process includes extruding the aluminum bar into profiles, aging, color anodizing or powder coating, packaging and finally shipping. You can see that all operations are batch. Customization is high. Many different profiles and many colors. How do I start?? Has anyone got lean experience with aluminum manufacturing?? I am now worried about the "technical" aspects of lean rather than "lean leadership" issues.
02/12/2007 09:02 PM
Print this message

Author Icon
JohnPod
John Podlasek



What are you trying to achieve?  What is the goal of this Lean initiative?
If you want someone to do it for you, hire a consultant.
You wont find a quick answer anywhere, especially on a website.

John
02/12/2007 09:02 PM
Print this message

Author Icon
774
Ranga Srinivas



Reynold,

How does your plant look like? Is it clean, neat, everything in its place? How many Extrusion presses you have? What are their OEE?
Select a Press which is making money. Start with 5S. Calculate OEE on that press. Your OEE shows where to start the improvements. If your press up-time is less than 80%, starting TPM program is the right way. Again the foundation for TPM is 5S.
During daily Production meeting ask why you did not meet that days production target?
Ask WHY 5 times.
Remember LEAN is a Journey not a destination.
By doing above I increased the OEE from 45% to 85% on several (Extruding) presses (7"-16"). in 18 months. If you need more information pl contact me at leansigma@gmail.com
Ranga S
02/13/2007 01:37 PM
Print this message

Author Icon
128947
Reynold Murphy



What we intend to do is much more than merely applying tools such as 5S and TPM. Certainly we will be using these tools as a foundation but we are striving for flow and pull. My question was that how can we create flow and pull in aluminum manufacturing considering that the process is a series of batch operations.

Yes, I will start with a current state value stream map but how would you suggest the future state map to be like based on the description of the process. The process includes extruding the aluminum bar into profiles, aging, color anodizing or powder coating, packaging and finally shipping
02/14/2007 08:46 AM
Print this message

Author Icon
JohnPod
John Podlasek



Reynold,
Does not matter right now if it is batch or not.  You can move batches, you already do. Just the process are not controlled or connected.  You can put in supermarkets. Kanban stations so that you do not overproduce. But you will have to stop production when Kanban is full.  You will need to spend some time figuring out what is the best Kanban size to balance your production.  Are you ready for this?
Once you get it all connected then you problems will arise and you will have to constantly keep tweeking due to the large batches and variation.  Start working to reduce batch sizes. Will take lots of time, and lots of effort.
I would be somewhat concerned if you are ready to do this. I don't like suggesting consultants, but you will be running blind.  A consultant may be what you need to get started properly, and help you with some hard decisions.  Putting in supermarkets, etc, may not help you to save a lot of money.  It may start to cause a lot of problems you are not ready for.  I am not sure.  You will find out.
 
02/14/2007 08:49 AM
Print this message

Author Icon
Shahrukh_Irani
Shahrukh Irani



You do have a continuous flow at the outset but also the worst of manufacturing systems .... the batch-type Process Flowline. :-( Does the technology and capital investment budget allow any form of multi-channelization within a process?  Setup reduction is easily something that you need.  How about allowing long runs of the Runners in the product mix, so that you do batch but at least do not hold all of that inventory for long?  At the production planning level, when you schedule orders, could you group by common profile (or other process similarities) within that period, so you reduce capacity lost to setups?  Could you cull some of your C-class items, and buy them from another source, then resell with your name on it?  Are you familiar with Group Technology?  I think this would greatly benefit you to understand (i) your business segments, and (ii) map those segments to setup reduction and job sequencing.  Check out a book by C.C. Gallagher and W. A. Knight .... that has been my bible for Lean in high-mix low-volume sectors.  
02/14/2007 02:44 PM
Print this message

Author Icon
DaveSimpson
Dave Simpson



Hi Reynold,

We are also a batch operation - we make frozen cakes, which are mixed in 600L batches, deposited into pans which are collected on racks of pans to be baked, then cooled (like your aging step), decorated, frozen and packaged. In our case, the changeover times for most product combinations are significant, and there's not much WIP as cake doesn't stay fresh long if it's not frozen.

We're a looooonnnnnnggg way from true flow, but we've started down the journey. So far we've been building capabilities - 5S, visual cues, improving machine effectiveness and availability, etc. We've defined product families based on manufacturing similarities, and changed the way we schedule the plant to reduce the variety of actual changeovers between these families (think of this as being like scheduling a rolling mill, where you start with wider product and progress through narrower product before finally stopping to re-grind the mill's rolls back to level.) Three people are learning to apply SMED techniques to our changeover process.

We've also changed the way we try to run the plant. Our constraint is usually the freezing operation, so we have developed buffers which allow us to keep them running through production breaks. No one works in the freezers, but there is a loader and unloader required - although these are normally part of the decorating and packaging operations. By offsetting break times for two people, and building a buffer between decorating and freezing (an anti-flow statement, I realize) we are able to generate 11% more output for the same labor content.

Finally, we have recently  begun TWI training, and are putting the finishing touches on a plan for a standardized work pilot, which will provide key corporate learnings for a roll out later this year. And we'll be making our first VSM map tomorrow, to see if this tool adds any value to our understanding of our processes and visibility to future improvements.

All of this without a clue about whether or how we'll actually achieve flow at the end of the journey. So far, we're up about 32% on output compared to 20 months ago, when we made our first small steps on this journey.

Perhaps we should form a smaller sub-group of batch based manufacturers to compare notes. Any interest?

Dave Simpson
Director of Operations,
The Original Cakerie
dsimpson@cakerie.com
02/14/2007 02:45 PM
Print this message

Author Icon
Larryod
Larry ODonnell



Is the over arching management goal to reduce inventory? Increase throughput? Improve customer service by reducing stock outs? Reduce non-value added activities? Improve flexibility? Reduce cycle time? Or just to do LEAN because you have been told to do it?

 

You are on the right track in you note you will start with a value stream map of the process.

 

 

However, could you add a little more flavor to your situation?

 

Are the final shipping sku's the same as the number of profiles extruded (not likely) or is some multiple of that number (x8).

 

What does the high level process tell you about the time quantity of batches queue, run, delay, cleaning, changeover or setup, etc? That is the cycle time for each high level process? Is extrusion the longest process? What does this tell you about pull? Are batches all of the same size? Do batches all take the same time? Is the lowest level of complexity just post extrusion (at the profile level)? What does this tell you about pull? Do any initial areas of investigation for improvement stand out? 

 

Are you experiencing customer fulfillment issues? What is your inventory turns? What are you scrap levels and where is quality lost?

 

Additionally you have said nothing about current perceived capacity - too much, too little, don't know? Globally satisfy need or do process bottleneck exist? Need more for future growth?  

 

And as always seek, at least, to ask to the level of the 5 whys.

02/20/2007 01:01 PM
Print this message

Author Icon
Robert_Simonis
Robert Simonis



RE: Aluminum operation

Reynold,

Larry's posting above as right on track.

Start with customer focus. What is your opportunity in the market?

If you make a variety of specialized alloys, the ability to have product to sell based on which alloys are in demand can be a significant advantage. If you are making just a few products, perhaps stock outs and order-to-delivery cycle is what is important to your customer(s). If you supply specialty material, perhaps quality is a market driver.

Determine your opportunity and it will focus your improvement efforts.

Take care.