Backflushing and inventory in an Engineer to Order Company

You are standing on shifting sand for sure. Backflushing to minimize making inventory transactions while knowing that your methods are inaccurate is not going to end well....
I think that the goal should be to 1) minimize the vulnerability for your purchasing team to make buying decisions on bad data, 2) Find the root cause of the method inaccuracies.

To minimize the potential of inaccurate purchasing decisions I would suggest keeping a controlled stocking location with manual/discrete purchase receipts going in and manual/discrete bulk material issues to a shop floor location as needed. It is not difficult to maintain a high degree of accuracy in your controlled stocking location with this method. Be sure to backflush from the shop floor location. Your buyers can then manage the known (accurate) inventory in the controlled stock location. All of your BOM inaccuracies, usage variances, etc. will show up in the shop floor backflush location. If the variances are large you can slice and dice and make multiple floor stocking/backflush locations. Assign some jobs to backflush from location A, other jobs from location B, etc. That way you might be able to identify which jobs are causing the largest variances and begin to investigate root cause and maintain those BOMs.

These are just a couple of ideas.
Good luck!
Dale

[cid:image001.jpg@01D1A1FC.C95F6DC0]
Dale Walker
Director, Continuous Improvement and Information Technology
Le Sueur Incorporated
ph.507.665.6204 x277
fx .507.665.8466
www.lesueurinc.com<http://www.lesueurinc.com>



[Non-text portions of this message have been removed]

We are running Epicor 10.0, but this topic applies to all versions. 


I'm curious about how other companies manage backflushing material.  Backflushing is causing complications for our company.  It would help us to know strategies in other companies that aren't selling a standard, stock product.


We are an engineer to order company, so our BOMs are not 100% accurate for our backflush items.  Our company decided to backflush select parts for two reasons.  First, it decreases and simplifies material transactions.  Second, some items are difficult to count ( Ex. Concrete and sand).


Our backflush material is stored in a non-nettable warehouse.  We store backflush parts in a non-nettable location to hide the quantity from purchasing. If purchasing depends on quantities that are inaccurate, then we risk over-buying or running out.


Since our BOMs aren't accurate, the quantity inaccuracy is very visible.  This leads executive management to ask for a physical inventory of our non-nettable inventory.  In Epicor 10, this is impossible.


Others in our company believe this is the only way to operate.  I'm wondering if people have found a way to successfully manage a similar situation.



So without all the visibility a company would normally have to material demands and inventory, how are you currently avoiding over stock or run outs and getting sufficient material on hand in time to support your manufacturing schedule?  How many inventory part numbers are you dealing with in this scenario?  Is cycle counting not an option should you choose to move these items to a nettable location?  What are the major obstacles to improving BOM accuracy?

 

Rob Bucek

Production Control Manager

D&S Manufacturing

301 E. Main St. | PO Box 279

Black River Falls, WI 54615

715-284-5376 Ext. 311

Mobile: 715-896-3119

rbucek@...

Visit our newly redesigned Website at www.dsmfg.com

 

From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com]
Sent: Thursday, April 28, 2016 8:21 AM
To: vantage@yahoogroups.com
Subject: [Vantage] Backflushing and inventory in an Engineer to Order Company

 

 

We are running Epicor 10.0, but this topic applies to all versions. 

 

I'm curious about how other companies manage backflushing material.  Backflushing is causing complications for our company.  It would help us to know strategies in other companies that aren't selling a standard, stock product.

 

We are an engineer to order company, so our BOMs are not 100% accurate for our backflush items.  Our company decided to backflush select parts for two reasons.  First, it decreases and simplifies material transactions.  Second, some items are difficult to count ( Ex. Concrete and sand).

 

Our backflush material is stored in a non-nettable warehouse.  We store backflush parts in a non-nettable location to hide the quantity from purchasing. If purchasing depends on quantities that are inaccurate, then we risk over-buying or running out.

 

Since our BOMs aren't accurate, the quantity inaccuracy is very visible.  This leads executive management to ask for a physical inventory of our non-nettable inventory.  In Epicor 10, this is impossible.

 

Others in our company believe this is the only way to operate.  I'm wondering if people have found a way to successfully manage a similar situation.

 

 

Thanks for replying.  Unfortunately, I can't control the decisions made at our company.  However, they come to me to help work out the issues.


How do we avoid over-stock and run-outs?  We don't avoid these issues.  We consider the quantities in this location to be unavailable and used.  Our company considers these locations to be like a Kanban.


How many inventory part numbers are you dealing with in this scenario?  We have about 700 parts in this scenario.


Is cycle counting not an option should you choose to move these items to a nettable location?  Yes, cycle counting would be an option.  However, our materials management department is un-willing to cycle count regularly.  The materials management department considers these parts to be productions problem and not theirs.


What are the major obstacles to improving BOM accuracy?  The first problem is specifications from the customer.  Most of our products are different every time. Many times we are engineering on the shopfloor.  Next, there isn't consistency in the method of production.  Different employees use different materials.  Finally, communication between engineering and production is failing for these parts.  When the parts are freely available, then there is no need to spend time correcting the drawing/BOM.


I know it's not an ideal situation.  I'm sure the answer is not going to be what our employees want to hear. 


Very interesting, I can appreciate the challenges you face, and there usually isn’t a simple answer to complex challenges such as you’re facing. Interestingly enough you listed many of the solutions to fixing your own issues, but as you said, no one wants to change, well you know the old adage, can’t keep doing the same things and expect different results. Most sound like they are well within your control exempting the ambiguity from the customer end, though it is possible with some effort to collaborate with the customers to help them improve their own internal processes so they can get you the necessary information you need. These types of efforts usually result in a win-win for both parties. It can be done, I’ve some experience with that. I think once you’re able to dial in your internal processes you’ll find that it’s much easier to manage these things within Epicor, after all your outputs will always mirror your inputs. Perhaps isolate the individual opportunities and start with one and work your way through them. Is there any way you can capture the dynamics at the operational level and use some of the toolsets to automate the updating of the BOM’s on the job to improve accuracy? We do that here as we have variation in material utilization on our main stream products and some engineered to order aspects to deal with as well.

Rob Bucek
Production Control Manager
D&S Manufacturing
301 E. Main St. | PO Box 279
Black River Falls, WI 54615
715-284-5376 Ext. 311
Mobile: 715-896-3119
rbucek@...<mailto:rbucek@...>
Visit our newly redesigned Website at www.dsmfg.com<http://www.dsmfg.com/>

From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com]
Sent: Thursday, April 28, 2016 10:28 AM
To: vantage@yahoogroups.com
Subject: RE: [Vantage] Backflushing and inventory in an Engineer to Order Company



Thanks for replying. Unfortunately, I can't control the decisions made at our company. However, they come to me to help work out the issues.



How do we avoid over-stock and run-outs? We don't avoid these issues. We consider the quantities in this location to be unavailable and used. Our company considers these locations to be like a Kanban.



How many inventory part numbers are you dealing with in this scenario? We have about 700 parts in this scenario.



Is cycle counting not an option should you choose to move these items to a nettable location? Yes, cycle counting would be an option. However, our materials management department is un-willing to cycle count regularly. The materials management department considers these parts to be productions problem and not theirs.



What are the major obstacles to improving BOM accuracy? The first problem is specifications from the customer. Most of our products are different every time. Many times we are engineering on the shopfloor. Next, there isn't consistency in the method of production. Different employees use different materials. Finally, communication between engineering and production is failing for these parts. When the parts are freely available, then there is no need to spend time correcting the drawing/BOM.



I know it's not an ideal situation. I'm sure the answer is not going to be what our employees want to hear.





[Non-text portions of this message have been removed]
>>Others in our company believe this is the only way to operate
>> I can't control the decisions made at our company. 
>> However, they come to me to help work out the issues.
This reminds of another site that was relatively new to Epicor.

They were used to the prior MRP system which was operated by a just a couple of people in the office.
No real-time transactions on the shop floor, no backflushing.
But... they did try to maintain BOMs for each order. Then simple MRP run and m
aterials were ordered and issued by batch processes.


>>We are an engineer to order company, so our BOMs are not 100% accurate for our backflush items
>> I'm sure the answer is not going to be what our employees want to hear.
After being on Epicor for a few months without making any changes in disciplines.... there were many problems with job costing and inventory.
When if finally got bad enough they assigned one person to verify BOMs and another to set and enforce shop practices. Not very popular but... I think you basically pay now or pay later, and it's more expensive later.

Understood.  Maybe we can run a tighter ship.  We have a few things

 

I would like to hear about how other companies use non-nettable locations and how they handle backflushing.  Are you successful?  Are there any challenges? 

 

Thanks,

 

Sharon Pulsifer

Fibrebond Corporation

318.371.6347

 

Hi Sharon,


We are also a  custom manufacturer and we use backflushing except for a select few items, but our BOM’s are 98% accurate when they go out to the floor.
In reading over your concerns and issues, in my opinion, you have a couple of choices.

 

A.       If your BOM’s aren’t accurate, you shouldn’t be backflushing.  It sounds like it’s difficult to have accurate BOM’s, due to the variables, so  I would either switch to manually issuing the material to the job OR

B.      Let backflushing do the majority of the work, but do regular (at least weekly) cycle counting to make any adjustments to the inventory of those materials.   And if you have to do adjustments that often, you might as

well just manually issue the material to the job and be done with it.  In addition, doing adjustments usually just goes to a variance account and the actual costs to the job are not going to be accurate.  I would assume

accurate job costs would be a factor in the decision as well.  It’s critical though to do the material issues in a timely manner (daily).

 

You have to be able to trust your inventory numbers in order to plan properly and to have accurate costs in your jobs, and an accurate value of your inventory.  It’s a tough battle to fight, but when folks start seeing the results,

they start to buy into the process.   Non-nettable seems like a band-aid solution and should be the exception, not the rule.  It’s better to experiment with safety stock, and/or min’s/max’s.  That gives everyone visibility to what’s out there,

the demands, requirements, and purchasing suggestions, etc.   The Time-Phase will be a more accurate depiction of what the needs are at that point too.  It won’t always be perfect – we still run into issues, but having an accurate inventory is                                               imperative. 

 

I would guess that most companies have gone through the pain you are experiencing at one time or another, and it’s not easy to get people to change their minds on how to manage those processes, but when they do and

they see the benefits and system starting to work for them, they will buy into it.

 

Good luck!

 

Julie Lehmann

Mercury Minnesota Inc.

 

 

From: vantage@yahoogroups.com [mailto:vantage@yahoogroups.com]
Sent: Friday, April 29, 2016 8:26 AM
To: vantage@yahoogroups.com
Subject: [Vantage] Re: Backflushing and inventory in an Engineer to Order Company

 

 

Understood.  Maybe we can run a tighter ship.  We have a few things

 

I would like to hear about how other companies use non-nettable locations and how they handle backflushing.  Are you successful?  Are there any challenges? 

 

Thanks,

 

Sharon Pulsifer

Fibrebond Corporation

318.371.6347