Sort Out Your Customer Orders

by Kien Leong

Demand Planning Inventory Allocation

How do you decide on demand allocation of inventory to customer orders?

In times of shortage risk, some manufacturers are tempted to hard allocate inventory.  The stock is held against the sales order upon order confirmation.

The inventory cannot be consumed for other orders.

All customers are born equal.  Yet some are “more equal than others”.  Every company has their customers who they place at a higher priority than others.

Isn’t hard allocation the best way to ensure that your most important customers get their product?

If you do hard allocate inventory at the point of order entry, does that eliminate the need to perform MRP runs for that order?

Short answers: Definitely not.  Yes, but at a huge cost to flexibility and very little benefit.

I will describe why, and then give you an Excel tool to download that proves it.

The tool is built on our simple Ten Minute Material Planner and it shows you how hard allocation leads to lower performance.  Follow this link to download and then return to read the rest of the article.

Demand Planning Service Levels

Best practices in demand planning calls for a service level to be allocated to customers.  This method will size stock and capacity to ensure that certain products and customers have a higher availability than others.

Calculating a service level depends on demand analytics, performed on large volumes of historical data.

Invariably, you’ll need a forecast and a way to define and measure forecast error.  There are some fundamental principles to forecasting, as I wrote about in an earlier article.

However, not all manufacturers can get meaningful results from forecasting at all.  Demand planning in medium- to high volume manufacturing is often easier than for high-mix/low volume production.

In a make-to-order business, it is more important to manage the orders you have on hand, rather than trying to reduce uncertainty and predict the future.

It is understandable that manufacturers look to a simple demand planning method.  A rule that is easy to understand and ensures the most important customers get their order filled on time.

When Does Hard Allocation Work and When Does It Fail?

Hard allocation of inventory holds aside stock against a customer order.  This is also sometimes known as “Hard Pegging”.

Soft allocation (sometimes “soft pegging”) can show the matching of demand with supply at the order level.  However, the inventory is not reserved against the order; it is just pegged and visible to the planner.  The planning process is free to peg it to a different sales order in a subsequent planning run.

Inventory is best managed with FIFO (first-in-first-out) principles.  Hard allocation violates FIFO, because a long lead-time order can consume inventory, even though there may be some demand and supply that comes in between.

Check the following example:

Hard allocation versus FIFO

Customer A orders on a long lead-time.  There is enough stock to fill the order by hard allocation, even though it is not due for another 4 weeks.

Customer B wants to place a smaller order on a short lead-time.  There is a supply order that will arrive in time to supply Customer A, but too late to supply customer B.

Hard allocation clearly reduces availability of stock.  If you hold aside the inventory that Customer A needs, then you are unable to supply Customer B.  Follow FIFO and keep the stock flexible: Both customers can be satisfied.

This phenomenon arises from a variability in lead-times.  If all customers worked to the same lead-time for the same product, then sales orders would be FIFO and hard allocation would still give you a FIFO inventory result.

Wouldn’t give you a better result than no hard allocation.  Doing about the same is the best you will get.

Furthermore, uniform lead-times are rarely something you can rely upon.

Demand Planning with Variable Customer Lead-Times

In times of plenty, everyone is happy with standard lead-times.  MRP systems hum away.  Orders are placed, but not chased.  We know straight away what is the best range at which to forecast.

In times of shortage, lead-times go out the window.  Even in these current economic times of supressed demand, companies are looking for any way to save cash.

Those days of inventory everywhere are going and they ain’t never coming back.

We might compile data in the item master and place lead-time settings against each part.  But if there is a shortage and an urgent customer requirement, do we place an order and wait patiently?  Er, no.

We get on the phone with the supplier and expedite.  Rush that order through, and maybe go looking for alternatives if they can’t satisfy our needs.  And we fully expect our customers to do the same to us.

Download the Example

This is an example with five customers that place orders against nine different products.  Of course, you can insert your own data in here.Demand Allocation - Excel Example

Each customer has a variable lead-time that is applied if the “Lead Time” is set to variable.  Otherwise it is an even 30 days.

Each run is going to be different, based on a random order quantity from a list of sales orders.  The menu sheet scores each run with 5 points for an on-time delivery, 3 points for a late delivery and zero points for an out of stock.

Before you download the example, please fill in this quick one-question survey:

Let us know how you get on.

 

{ 13 comments… read them below or add one }

Tony Rice October 20, 2011 at 11:46 pm

Makes perfect sense to me Kien. To add to what you are saying – one of our clients wanted to make sure their “1st class” customers got preferential supply. The answer – once you are down to say 20% of the safety stock level, tell your 2nd class customers that you are out of stock.

S Egner October 24, 2011 at 4:20 am

Thanks a thoughtful look at this.
It might make sense for low volume materials on project manufacturing. For mass production it ties your hands and prevents you accepting short lead time orders.

JP October 24, 2011 at 8:04 pm

great blog! we use hard allocation to protect our customers- those ones that are “more equal” than others. your tool shows better service over all for soft alocation, but if an OTD for a VIP customer scores more than an OTD for a normal customer then hard allocation could still beat soft.

Dave George October 25, 2011 at 12:54 am

We do sometimes use hard pegging in Oracle. We are trying to move away from this as it does make last minute inquiries harder to please

Nigel Pitts-Drake October 31, 2011 at 7:17 am

A very interesting read. We often “ring fence” products for our more favoured/higher margin customers and when push comes to shove, they are first in the pecking order.

John Wickham October 31, 2011 at 7:40 am

Look after your longer serving customers and they will look after you.

Gabriele Tettamanzi October 31, 2011 at 8:21 am

Great article, thank you. We could also flexibly allocate production to customer order,at least the capacity of finished products lines. Gabriele

Surika Prinsloo October 31, 2011 at 8:26 am

Great article and another excellent tool to use. Thanks!

andrey andrey October 31, 2011 at 10:31 am

nice an useful article!

hatice uras October 8, 2012 at 12:29 pm

güzel… emeğinize sağlık…

Deepak G Pardeshi January 24, 2013 at 10:36 pm

Dear All,
As per your tool, we have no. of SKU”s available with us,We have no of customer also & No. of orders are coming. But the orders are not serve on FIFO basis. Please could you please explain me to Tackle this situation. We have track the details with datewise also.

Vaibhav PATIL March 14, 2013 at 9:57 am

very nice usefully article

sasithorn tee June 1, 2013 at 6:37 am

thanks!

{ 40 trackbacks }

Previous post:

Next post: