**Globalization**

Remember how Americans feared that jobs were going abroad? It was a fear that production line workers were going to be outsourced to countries who could pay a worker for one day abroad what a US based worker cost an hour. This really wasn't a time for a depression -- it was a time for America to embrace an opportunity. Here's why: when ordering in bulk transcontinentially, it must travel by air or sea. The cost for air is 4X that of sea. However the ocean time is 4-6 weeks (including customs). So, companies were faced with a delimma of bulk ordering and tying up large sums of money vs. staying stateside. We're passed that now. The production jobs went abroad, but if the production line workers were interested in re-training, there were opportunities in managing the inventory. Getting the right inventory to the right place and in the right form (assembled but not broken). Additionally because of the huge crates on the ocean liners, more of the product needed to be tested for quality before it reached the consumers. Then there had to be people who were going to keep track of the rejects. At which point, many managers made the decision to repair in-house and then charge-back to the overseas division. Clever solutions. It's solutions like that, though, that can take a company who's moving one type of job abroad and creating another job stateside. Timing and open-mindedness with sharp, talented employees do it though.

Speaking of, there is a decision when to say, "We have our largest client on line 1 asking when the product is going to be here." and you have one person who is controlling that relationship. That person needs to know that it is worth changing the distribution channels from ocean to air. Another job saved.

Basically, America's production line specialist transitioned into warehouse specialists. They became practical learners of industrial engineering. That is how some jobs have been saved in a globalization.

**Seasonality**

Are you frustrated with fluctuating inventory levels? You are carefully watching the purchasing department’s budget. Cash flow is of essence because you’re looking to expand into a new technology and you would like to finance it with the fewest possible long term notes. One wrong purchase of raw materials could tie up cash flow for additional purchases that cannot be used for months.

It’s the end of the quarter, and the accounting department has reminded you that it needs inventory levels for write-offs due to shrinkage. To top it off, you are paying overtime for employees to count inventory. If you just had less inventory build-up of seasonal items to count…

You’ve looked at sales, specifically your seasonal product lines, and they aren’t as promising as you’d like. A recent visit through your warehouses shows that you’re paying for space that isn’t occupied. Senior management reports that consolidating warehouse space is an option. They also advise that sub-leasing the space could follow. You have the burning desire to challenge them with the “how did we get this build-up of seasonal inventory, which doesn’t have as high a yield on a secondary market, in the first place?” A down market isn’t the answer. Your competitors with seasonal products don’t have the warehouse real estate that you do but they are surviving. If you could just calculate the most accurate forecast…

**Seasonal vs. Cyclical**

Seasonal products recur. Lawnmowers, bathing suits, parkas, outside furniture are examples of such seasonal lines. Often because of fashion and/or technology improvements, the last season’s product is not re-used, or must be heavily discounted.

Products that have cycles return often, in the same form, but after they have undergone a process to put them back into their marketable form. One-use cameras that are turned in and reloaded with film. They originated from the same manufacturing plant. The ready-for-sale product exits one door of the plant. The cameras with no more film enter in another door. The cycle repeats itself.

**Remedy**

Here are two examples for when Winters Law is desirable. If your inventory looks like your EKG, then Winters Law is not an option; it is a necessity. If you have items with extraordinarily long shelf life, then let’s deploy Winters Law. The starting point is the same as it would be if you were visiting a cardiologist. Historical data.

One of the most fascinating, and unarguably the most complex, is seasonality. So, by, seasonality, we're suggesting some level of deterministic modeling. That means we know the inventory is not constant but every period, it climbs and then falls again. Coincidentially, the layman solution is Winters Law (1960). Through a series of iterations, or "smoothing," we calculate a common factor, or a more technical term, an exponent. Before we get into Peter R. Winters' work, let's talk about the terminology before we begin.

We need to know the "level." What is the [estimated] count of the material (or product)? This conveniently is referred to as *a*.

Following *a* is *b*. *b* is the trend. This is **_not_**seasonality. This is illustrated by thinking of the IT world. First there's a boost in hardware. Then, once the hardware becomes faster and bigger, software programmers go crazy at the keyboard. There's a huge leap in software. Every four months, we now expect the latest changes in software. Every December, though, there are huge purchases for the latest new iPOD, game, laptop - a plethera of gadgets. The trend *b* is the every four months. The "December" spike is our next term.

Seasonal index. F_{t}.

Of course "t" is the period.

"P" is the period for the entire season.

Next bring in the Vice President of Sales. Sales is the origin because the department head sets the sales standard. This individual is closest to the demand, which we call x.

Subscripting this x refers to the period for that demand. For example, if your product is reviewed monthly, then there are time intervals, t, in one period (year). Thus, there are twelve demands, x_{1}, x_{2}, x_{3},.. x_{12}. Sum (∑) the x_{1}.. x_{12} and divide by 12 to reach the mean, or the average, . To calculate a moving average, you simply take the demands for the past 12 intervals. Example, for January, use January – December. For February, use February – January, and so on. To use supply chain geek language, this is called the simple moving averages.

Are simple moving averages enough? Not quite, especially if the inventory resembles an erratic EKG. An erratic EKG has a great deal of “white noise,” and to compensate, Your Business is *Alive!* takes a secondary moving average. The sales VP needs to come prepared. The more data this important person brings to the meeting, the better YBIA! can help your company plan for seasonality.

There is moving average #1 and moving average #2. Add them together and divide by 2. This is known as a centered moving average. If you have 12 intervals period P, then it is a called a centered 12 point moving average.

F is denoted as the seasonal factor. Subscripting it with an interval, t, shows the seasonal factor for that interval within the period. Conclusively, that is F_{1}…F_{12} for the entire period, P. To calculate an F_{t}, take the demand, x_{t}, and divide by the center moving average_{t}.

We’re only doing this exercise to find out what the seasonal indices are, in this example, for each month. If your sales VP came with all of the data, then now it’s time to sum all of the January F_{t} and divide by the number of F_{t}s. In other words, get the average of F_{t} where all t = January. This index is now referred to as F_{i}.

Normalize your F values. To normalize, sum all F values. P divided by this value * F_{i}..and so on. If you do not do this, your trend line and level cannot be adjusted. Therefore, this defeats the purpose of creating a seasonality index.

If we were to pull back all the noise and all the seasonality, a level exists. It’s a horizontal line, which we will call the level and label it *a*. It’s holistic in nature because if we knew *a*, then we wouldn’t have a need for warehouses of excess inventory, for example.

We take each x_{1} / F_{1} to give us each estimate of a_{i}.

Plot these a_{i} values to the best fit line, also known as a regression line. The slope is referred to as b. (Yes, algebra is coming back to haunt us again!) It is commonly known as the trend line.

Now, the sales VP is exhausted from providing all of the historical data. Hang in with us! We have all the data we need to provide a seasonal indexed forecast. The line (so far) is the trend line. It begins at a historical point and ends at our current point in time. If we were to continue drawing it, we begin to see what the forecast is going to be. (All subscripts were removed for visual simplicity.)

x = a + b * F

You know F_{i}, depending on the interval, it varies. You know a and b. There’s nothing left but x. Remember, *a* and *b* are going to stay constant. It’s the F and x values that are changing. Intuitively they need to [change], because we’ve now successfully accounted for the seasonality. You should have a basic understanding and a graph to illustrate on a high level why the numbers provided are crucially important to addressing seasonality.

These sales forecasts are subsequently provided to the Vice President of Manufacturing. Contact Your Business is *Alive!* and we can show you how to do this for each of your product lines. You don’t have to understand the math because we do.

__Note__: Smoothing constants are used for trend lines. Smoothing constants range between 0 and 1. Smoothing constants are used for the *a*, *b* and F. The older the data and/or the more erratic the data, the higher the constant. Explaining this much via the website would be unnecessarily confusing unless you are actually going to do it. Your Business is *Alive!* will do it for you! Contact us today!

Special thanks to Dr. Holly Lutze, University of Texas at Dallas who was my mentor while I was earning my degree.

**Inventory Replenishment Policies**

Both retail and manufacturing companies are concerned with inventory levels. In lieu of people, the plants receive in, handle, and ship out thousands of dollars worth of inventory daily. However, manufacturing plants use materials requirement planning (MRP) systems. They do not use inventory replenish systems when designing their products. Simply understood, inventory replenishment policies are ideal when there can be partial inventory sold. A convenience store can sell milk even though it has a stock-out of cereal. If you are not a manufacturing firm, read on. If you are, then click to MRP.

**Reorder Point vs. Interval Order Comparison**

Specifically, Your Business is *Alive!* asks what is your replenishment policy? A continuous review system tracks the level of inventory and once it drops below a certain point, an automatic order occurs. In order to best help, Your Business is *Alive!* examines how your company determined its reorder point. The reorder point is the number that you’ve established as the quantity at which it is time to place an order. If you wait, for example, until you are completely depleted of raw materials, then you are risking a stock-out. A stock-out is when your company cannot meet the demand of any one of your customer orders with the right product, in the right condition, at the right time. We all know the risk of stock-outs – an opportunity for valuable customers to go elsewhere or for us to discount the product once it does arrive to compensate for the inconvenience of a stock-out so severely, it erodes us of profits, or worse, falls below the break-even point. The customer can be lured back in when there is stock-out if it is discounted and/or if it is “rush delivered.” This practice eats profits. (Unless it has to be today, many of us will easily wait for a product if we can reap a savings!)

The difference between the two types of reorder point policies is the quantity that is ordered. If the continuous policy is in place, once the reorder point is reached, a fixed quantity is ordered. For example, if it is time to order, the ROP = 600 and the inventory level is 590, the amount quantity lot size ordered is 50. Therefore, 50 is ordered.

In the interval ordering, also known as periodic review, the quantity that would have been ordered in the above example is 10. It is the difference between inventory on-hand and the ROP. On the surface, one would think that the latter example is preferred -- order only what you need. However, your vendor may not allow small quantities to be ordered. Or, your company is in a rural area, and you have a minimum $ amount that you must make. You know that this is a product that has a higher chance to have an inventory turn, therefore, you've opted to always order it. Bulk quantity order sometimes prove to be cost effective. See EOQ (economic order quantity) or refer to the EOQ Webinar.

Safety stock is the quantity that kept on hand, is smaller than the reorder point quantity. Ideally, after the fixed order quantity is made, the suppliers deliver it prior to your company’s safety stock level. The difference in safety stock vs. reorder point is the lead time. Obviously, safety stock is only designed to be used in emergencies. If using safety stock becomes part of your daily operations, then the probability of stock-out increases and the service level is likely to decrease. The service level is expressed in terms of percent, and the percent of stock-out total 100%.

The safety stock factor is denoted as "z" in a normal distribution. "z" is the link between the aforementioned fixed order quantity model and the following fixed interval ordering policy. The fixed interval order policy says that regardless of current inventory levels, re-order enough inventory to bring inventory up to a specific level. Your company does not rely on an economic order quantity but rather on an ideal inventory level. (Remember economic order quantities are calculated based on the trade off of inventory carrying costs vs. order costs.) The "z" factor is the safety factor and if we know what the ultimate safety stock quantity is, then we can determine "z." (This relates back to our statistics classes. Remember the good ol’ normal distribution? This is the "z" we’re referring to.) The safety stock quantity is established at your company because your company advertises (or can advertise), for example, as maintaining “a service level greater than XX%.”

Let’s compare the formulas you need to know to establish ideal reorder quantities or if your company uses interval ordering, the target inventory level. Before we begin this, below is the checklist needed:

calculate your economic order quantity;

know the demand; and

>compute your safety stock.

Contact Your Business is *Alive!* today for assistance in this process and establishment of Inventory Replenishment Policy.

Concept Developed by: Professor Roger G. Schroeder, University of Minnesota