Inventory Management Review

What to do about seasonality

Without a fairly accurate sales forecast, managing inventory is essentially impossible. There is no way to determine optimal inventory levels and a corresponding production schedule without a projection of sales. Deciding on quantity of shipments received from suppliers is also out of the questions without a forecast. Once the forecast lands on your desk, you can begin to do your work. The first thing you should notice regarding the sales forecast is whether or not the product is seasonal or not. Assuming it is seasonal, there are several issues to consider regarding its production schedule. Ultimately, you need to decide whether or not you want to produce the product at a constant level throughout the year (thus building up inventories and selling them off during the busy season) or whether or not you want to produce products at a level that chases demand (producing the items right before you can sell them).

The first issue to look at is the holding cost of your product.  If holding costs are very low relative to the revenue received per unit, it may be worthwhile to produce at a constant level. Even if you only sell the item during the fall, production in winter that is held as inventory during spring summer may be worthwhile if holding costs are low enough. Chances are, if your product is that seasonal, it is not worthwhile to produce at a constant level.

Another issue to look at is how chasing demand may affect your production schedule. If you decide to produce only during four months of the year in order to accommodate the four months worth of seasonal demand that you have, you should consider how intensive those four months will be. If your company produces a plethora of other products, then this may be a simple reallocation of resources. Ideally you would have another product that just so happens to be sold during the remaining seasons. A good example of this is in the lawn mower industry. Unsure with what to do with themselves during the winter season, they began to produce snow blowers, which is a perfect fit as it is a similar design with an opposite seasonality…personally, I feel they should have just sold to the southern hemisphere, but then again, I’m not too sure about how many lawns there are that needing mowing in Paraguay. If however this is the only product your company makes, then there should be some serious considerations regarding the idle time of company resources. Perhaps outsourcing will become an appealing choice once you realize that without outsourcing, you will have 8 months of idle capacity.

Another important factor in determining the choice of production schedules is how your suppliers will respond. Perhaps they will prefer that you produce all at once so you can have large quantities sent to you which may result in quantity discounts. However, it is also possible that your supplier would prefer to produce at a constant level.

These are just some of the factors that will help you determine what do when facing the management of a seasonal product.

Choosing your location

Deciding where to put your warehouse, factory, and offices can become complex, but is a crucial step in managing your operations and in helping you cut down your inventory costs. First I will discuss its importance and then the steps involved.

The most important reason for choosing a good location is to cut down on shipping costs. It is essential to minimize both inbound and outbound shipping costs throughout the entire supply chain when determining your ideal location. This reason should be fairly straightforward. Setting up a factory in the same state where the warehouse, distributor and bulk of the sales come from would save a company an incredible amount of money. It is also important to try obtain parts from suppliers that are located in close proximity to your in-house operations.

As far as inventory management is concerned, inbound and outbound logistics can be greatly affected by having a prime location. The reason for this is because of lead times. In a recent article, I discussed the Re-Order Point. A large reason inventory is kept is because of uncertainty in demand. If demand rises above sales forecasts expectations, inventory is in place to help fulfill orders until more parts can arrive. With today’s overnight shipping, this may not seem like a big deal, but a location close to suppliers can help facilitate shorter lead times.

A less objective advantage to consider is regarding whether or not you really want to travel so far every time you have to go to one of the locations involved in your supply chain. Objectively speaking, travel costs can really begin to add up.

In order to figure out your location, it is essential to determine where all of your inbound parts are coming from, and also where all of your outbound goods will be shipped to. You will also need to determine how many of each good you will be shipping or receiving and how much each weighs. Ultimately, you need to take a weighted average of shipping costs per mile and then use the following formula to determine your optimal selection. More specifically you need to treat each mile traveled as a cost per mile and then take the global latitude in which they are traveling from and use it find an average, and then do the same for the longitude.

For example, if you have a supplier that charges $.0010 per mile, per good and is located at a latitude of 40 degrees and a longitude 100 degrees. Assume you also have a distributor and the cost to ship to him is $.0015 per mile per good and the longitude is 10 degrees and the latitude is 80 degrees. Both locations are in the northwest hemispheres. Annually you need to ship out 100,000 units and need to ship in 200,000 units.

Ideally the center of gravity formula results in an ideal longitude of 20.5 degrees N, and ideal latitude of 75.3 degrees W. Because the formulas are a bit more complex than previous ones discussed on the site, I have made an excel spreadsheet covering these formulas with the example solved.

These ideal values are only suggestions. State taxes, state wages, available land, labor, and other resources should also be taken into consideration. The center of gravity should be used specifically as a way to find a location close to where you should be.

Radio Frequency Identification Tags

In the last article, I discussed the how to calculate the Re-Order Point (ROP).  To recap, the importance of the ROP is that it the mathematically ideal point at which to re-order a part in order to minimize holding costs while preventing a stock out at a certain service level.  To help calculate the ROP, I have added an excel spreadsheet with sample data.  Feel free to download this file and change the data on the two sheets in order to calculate your own ROPs.

I hope by now you understand why the ROP is important and how to calculate it.  The problem now is how to know when you’ve reached the pre-determined point at which you want to re-order.  Before I discuss Radio Frequency Identification Tags (RFID), I am going to cover some basic practices of inventory counting practices.

One way to keep track of inventory is to determine how many you have ordered and how may you have used.  The remaining difference should be what you have in inventory.  To make this process more accurate, many companies put barcodes on items and scan them at all locations to help track them.  Unfortunately, for various reasons, this is not always as accurate is needs to be.  Whether it is because of scrap, or stolen goods, this method of inventory counting is rarely 100% accurate.

Many organizations prefer to use a physical count of their materials.  This means that some poor sap has the job of going around and counting exactly (or inexactly) how many of each material there is.  This is a great way to manage inventory if you’re a handmade violin manufacturer and you make 1 violin a week.  If however, you run an audio accessory plant that produces over 10,000 pairs of headphones a day, then a physical count might not be so simple.  Or, imagine if you had not only 1 product, but had over 1000 and maintaining optimal inventory levels was an essential part of your business.

Wal-Mart has over 1000 products on their shelves and maintaining optimal inventory levels is essential to their business.  One of the reasons they are able to keep costs so low is because of their high inventory turnover which is made possible by maintaining particular inventory levels.  A physical count for a company with as many products as Wal-Mart is out of the question.  Wal-Mart employs RFID tags in order to keep track of their inventory.

These tags send out radio frequencies tracked by computers which allow managers to keep track of where goods are at any point in time.  These tags allow managers to not only track how much of a product is in their factory at a given point in time, but also where it is.  RFID tags are essentially a manner of making the goods tell a computer where they are allowing the computer to count the goods, thus eliminating the need to count inventory physically while still keeping accurate counts of the inventory.

Currently, IBM is working on a more affordable RFID system.

Read More about RFID:  http://www.rfidgazette.org/

Safety Stock

First of all, here's the formula so you don't have to dig through my well-written article for it.

Safety Stock:  {Z*SQRT(Avg. Lead Time*Standard Deviation of Demand^2 + Avg. Demand^2*Standard Deviation of Lead Time^2}

If that wasn't clear to you, I suggest reading on.  This article will explain in detail what safety is used for, and how to use it.

Inventory management is about two things: not running out, and not having too much. Our desire to not run out, along with uncertainties in demand and supplier lead times are why we have inventory in the first place. Essentially, inventory is a reserve system to prevent a stock out. However, as important as it is to prevent such a stock out, we also don’t want to hold onto too much inventory because of holding costs. So how do you balance the two and what is the right amount? More importantly, when should you re-order in order to prevent a stock out? The answer to this can be determined by obtaining and applying the following information about the inventory you wish to manage.

Re-order Point (ROP)

1. What is the average lead time for the part/finished good that you need?

2. What is the standard deviation of that lead time? It is very important to track how long shipments take from you suppliers. If you are not doing this, start. It should be your top priority. Assuming you have tracked the data, excel can very easily help you determine your standard deviation. In excel, go to the toolbar and click on Insert, then click on Function, and choose STDEV and click ok. Then, enter in as much lead time data you have and presto, you have your standard deviation.

3. What is the expected demand you are working with?

4. What is the standard deviation on this demand?  Perhaps this is something you will be familiar with from experience, however, if not, this is something you should be able to squeeze out of Ted from the marketing department.  One way to find it is to look at historical demand and use the STDEV function in excel to determine it.

5. How sure do you want to be that you aren’t going to run out? 90%, 95%, 98%, 99%? Whatever you decide, this will become your service level. Using this percentage, a statistical z-table should be used to get the corresponding “z-value.” A good z-value webpage can be found at http://www.inventoryops.com/safety_stock.htm. So, for example, if you want a 98% service level, you would use 2.05 as your z-value.

Ok, so you’ve gathered this data, now here’s what you do with it.

(Underlined section is safety stock)

Re-order point=Average Lead Time*Average Demand + Z*SQRT(Avg. Lead Time*Standard Deviation of Demand^2 + Avg. Demand^2*Standard Deviation of Lead Time^2)

In this formula, the first term (Average Lead Time*Average Demand) is the average demand.

The second term {Z*SQRT(Avg. Lead Time*Standard Deviation of Demand^2 + Avg. Demand^2*Standard Deviation of Lead Time^2} is the term that allows for the safety stock. In other words, the second term is the optimal safety stock level.

It is not simple to gather all the data that is needed for the calculations. For a product with multiple parts, each part needs to have its own re-order point calculations and its own safety stock calculation. This can all become very confusing if proper computer modeling is not employed.

Although I mentioned excel earlier, excel is probably not sufficient for your company’s software needs. If you have not already done so, it is very important to look into an integrated software package for these calculations and many others.