MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 1 of 30

FUNDAMENTALS OF INVENTORY
Within most organizations inventory exists in a variety of places, and in a variety of forms, and for a variety of reasons. Although these inventories represent a substantial cost investment (in some cases as much as 50% of total capital invested), they are necessary to provide a desired level of service to customers. The objective of inventory management is to strike a balance between inventory investment and customer service.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 2 of 30

FUNCTIONS OF INVENTORY
Inventory exists for a variety of reasons (i.e., serves several functions) within organizations. 1. Decoupling stages in the production process. Inventory between successive stages of a transformation process make each stage less dependent upon the output of the prior stage. If there is an interruption in output at one stage, succeeding stages may be able to continue operation by feeding off the inventory held between stages. This applies both to internal operations and to external linkages with suppliers. This inventory is called buffering inventory. 2. Decoupling from demand fluctuations. This manifests itself in both seasonal inventory and safety stock. When there is predictable variation in demand throughout the year, and when an organization does not have the capacity to produce peak demand when it is demanded, the organization may have to produce and store finished products in advance of that demand. This inventory is called seasonal inventory. When there is unpredictable (i.e. erratic and random) short term variation in demand, the organization may have to maintain additional inventory to cover the unpredictable spikes in demand. This inventory is called safety stock. 3. Volume purchasing. Purchases in large quantities may result in reduced purchase price and/or reduced delivery cost. Such incentives often lead organizations to acquire more inventory than is immediately needed. This inventory is called volume discount inventory. 4. Hedge against possible future events. In many instances organizations perceive that there may be a disruptive economic or environmental event in the not too distant future. Inflation may suggest that there will soon be a price increase in some supply. Labor negotiations may suggest that an impending trucker strike might affect delivery of supplies. Weather conditions indicate that a brewing tropical storm might affect shipments of supplies. In circumstances like these organizations may choose to order more inventory than is immediately needed to provide protection in the event that any of these situations actually occur. This inventory is called hedge inventory.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 3 of 30

TYPES OF INVENTORY
To better accommodate the functions of inventory, organizations maintain four types of inventories. 1. Raw material inventory. Materials that are usually purchased and have not yet entered the transformation process. 2. Work-in-process (WIP) inventory. Materials and components that have undergone some change but have not yet advanced to the stage of completed product. 3. Finished-goods inventory. Completed products awaiting shipment. 4. Maintenance/repair/operating (MRO) inventory. Supplies necessary to keep machinery, processes, facilities, and office operations running. These items do not get absorbed into the products being made, but are crucial to the smooth operation of the organization. They range from such things as lubricating oil for machines and janitorial cleaning products, to printer toner cartridges and other office supplies.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 4 of 30

BASIC INVENTORY DECISIONS

There are two basic decisions that must be made for every item that is maintained in inventory. These decisions have to do with the timing of orders for the item and the size of orders for the item. We will be examining several models and philosophies related to these two decisions. As noted on page 1, the objective of these inventory management models is to strike a balance between inventory investment and customer service.

Basic Inventory Decisions

How Much? Lot sizing decision Determination of the quantity to be ordered.

When? Lot timing decision Determination of the timing for the orders

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 5 of 30

INDEPENDENT VS. DEPENDENT DEMAND INVENTORY

Before examining specific inventory models, an important distinction must be made. Some inventory items can be classified as independent demand items, and some can be classified as dependent demand items. We need to make the timing and sizing decisions for all inventory items, but we will find that the manner in which we make these decisions will differ depending upon whether the item has independent demand or dependent demand. Independent demand inventory item: Inventory item whose demand is not related to (or dependent upon) some higher level item. Demand for such items is usually thought of as forecasted demand. Independent demand inventory items are usually thought of as finished products. Dependent demand inventory item: Inventory item whose demand is related to (or dependent upon) some higher level item. Demand for such items is usually thought of as derived demand. Dependent demand inventory items are usually thought of as the materials, parts, components, and assemblies that make up the finished product.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 6 of 30

RELEVANT INVENTORY COSTS

Relevant Inventory Costs

Item Costs
Direct cost for getting an item. Purchase cost for outside orders, manufacturing cost for internal orders.

Holding Costs
Costs associated with carrying items in inventory. Storage and other related costs.

Ordering/Setup Costs
Fixed costs associated with placing an order (either an ordering cost for outside orders, or a setup cost for internal orders).

Shortage Costs
Costs associated with not having enough inventory to meet demand.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 7 of 30

BEHAVIOR OF COSTS FOR DIFFERENT INVENTORY DECISIONS

When assessing the cost effectiveness of an inventory policy, it is helpful to measure the total inventory costs that will be incurred during some reference period of time. Most frequently, that time interval used for comparing costs is one year. Over that span of time, there will be a certain need, or demand, or requirement for each inventory item. In that context, the following describes how the annual costs in each of the four categories will vary with changes in the inventory lot sizing decision. Item costs: How the per unit item cost is measured depends upon whether the item is one that is obtained from an external source of supply, or is one that is manufactured internally. For items that are ordered from external sources, the per unit item cost is predominantly the purchase price paid for the item. On some occasions this cost may also include some additional charges, like inbound transportation cost, duties, or insurance. For items that are obtained from internal sources, the per unit item cost is composed of the labor and material costs that went into its production, and any factory overhead that might be allocated to the item. In many instances the item cost is a constant, and is not affected by the lot sizing decision. In those cases, the total annual item cost will be unaffected by the order size. Regardless of the order size (which impacts how many times we choose to order that item over the course of the year), our total annual acquisitions will equal the total annual need. Acquiring that total number of units at the constant cost per unit will yield the same total annual cost. (This situation would be somewhat different if we introduced the possibility of quantity discounts. We will consider that later.)

Cost

Total Annual Item Cost (assumes no quantity discounts)

Lot Size (how much decision)

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 8 of 30

INVENTORY COST BEHAVIOR (CONTINUED)

Holding costs (also called carrying costs): Any items that are held in inventory will incur a cost for their storage. This cost will be comprised of a variety of components. One obvious cost would be the cost of the storage facility (warehouse space charges and utility charges, cost of material handlers and material handling equipment in the warehouse). In addition to that, there are some other, more subtle expenses that add to the holding cost. These include such things as insurance on the held inventory; taxes on the held inventory; damage to, theft of, deterioration of, or obsolescence of the held items, and opportunity costs associated with having money tied up in inventory. The order size decision impacts the average level of inventory that must be carried. If smaller quantities are ordered, on average there will be fewer units being held in inventory, resulting in lower annual inventory holding costs. If larger quantities are ordered, on average there will be more units being held in inventory, resulting in higher annual inventory holding costs.

Cost

Total Annual Holding Cost

Lot Size (how much decision)

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 9 of 30

INVENTORY COST BEHAVIOR (CONTINUED)

Ordering (or setup) costs: Any time inventory items are ordered, there is a fixed cost associated with placing that order. When items are ordered from an outside source of supply, that cost reflects the cost of the clerical work to prepare, release, monitor, and receive the order. This cost is considered to be constant regardless of the size of the order. When items are to be manufactured internally, the order cost reflects the setup costs necessary to prepare the equipment for the manufacture of that order. Once again, this cost is constant regardless of how many items are eventually manufactured in the batch. If one increases the size of the orders for a particular inventory item, fewer of those orders will have to be placed during the course of the year, hence the total annual cost of placing orders will decline.

Cost

Total Annual Ordering Cost

Lot Size (how much decision)

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 10 of 30

INVENTORY COST BEHAVIOR (CONTINUED)

Shortage costs: Companies incur shortage costs whenever demand for an item exceeds the available inventory. These shortage costs can manifest themselves in the form of lost sales, loss of good will, customer irritation, backorder and expediting charges, etc. Companies are less likely to experience shortages if they have high levels of inventory, and are more likely to experience shortages if they have low levels of inventory. The order size decision directly impacts the average level of inventory. Larger orders mean more items are being acquired than are immediately needed, so the excess will go into inventory. Hence, smaller order quantities lead to lower levels of inventory, and correspondingly a higher likelihood of shortages and their associated shortage costs. Larger order quantities lead to higher levels of inventory, and correspondingly a lower likelihood of shortages and their associated costs. The bottom line is this: larger order sizes will lead to lower annual shortage costs.

Cost

Total Annual Shortage Cost

Lot Size (how much decision)

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 11 of 30

INVENTORY COST BEHAVIOR (CONTINUED)

All Four Cost Categories Combined: When all four inventory cost categories are superimposed on the same graph, we obtain the following (somewhat cluttered) picture which suggests that there is one best answer to the how much decision. The quantity that should be ordered is the lot size that corresponds to the lowest point on the total annual cost curve. This quantity is referred to as the economic order quantity, or EOQ.

Cost

Total Annual Cost

Annual Holding Cost Annual Shortage Cost Annual Item Cost Lot Size (how much decision) Annual Ordering Cost

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 12 of 30

BASIC ECONOMIC ORDER QUANTITY (EOQ) MODEL

The EOQ model is a technique for determining the best answers to the how much and when questions. It is based on the premise that there is an optimal order size that will yield the lowest possible value of the total inventory cost. There are several assumptions regarding the behavior of the inventory item that are central to the development of the model EOQ assumptions: 1. Demand for the item is known and constant. 2. Lead time is known and constant. (Lead time is the amount of time that elapses between when the order is placed and when it is received.) 3. When an order is received, all the items ordered arrive at once (instantaneous replenishment). 4. The cost of all units ordered is the same, regardless of the quantity ordered (no quantity discounts). 5. Ordering costs are known and constant (the cost to place an order is always the same, regardless of the quantity ordered). 6. Since there is certainty with respect to the demand rate and the lead time, orders can be timed to arrive just when we would have run out. Consequently the model assumes that there will be no shortages. Based on the above assumptions, there are only two costs that will vary with changes in the order quantity, (1) the total annual ordering cost and (2) the total annual holding cost. Shortage cost can be ignored because of assumption 6. Furthermore, since the cost per unit of all items ordered is the same, the total annual item cost will be a constant and will not be affected by the order quantity. Inventory levels will fluctuate over time as in the following graph: Inventory Level Q Q Q Q = Order Size Q/2

Time EOQ symbols: D = annual demand (units per year) S = cost per order (dollars per order) H = holding cost per unit per year (dollars to carry one unit in inventory for one year) Q = order quantity

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 13 of 30

CLASSIC ECONOMIC ORDER QUANTITY (EOQ) MODEL

We saw on the previous page that the only costs that need to be considered for the EOQ model are the total annual ordering costs and the total annual holding costs. These can be quantified as follows: Annual Ordering Cost The annual cost of ordering is simply the number of orders placed per year times the cost of placing an order. The number of orders placed per year is a function of the order size. Bigger orders means fewer orders per year, while smaller orders means more orders per year. In general, the number of orders placed per year will be the total annual demand divided by the size of the orders. In short, Total Annual Ordering Cost = (D/Q)S Annual Holding Cost The annual cost of holding inventory is a bit trickier. If there was a constant level of inventory in the warehouse throughout the year, we could simply multiply that constant inventory level by the cost to carry a unit in inventory for a year. Unfortunately the inventory level is not constant throughout the year, but is instead constantly changing. It is at its maximum value (which is the order quantity, Q) when a new batch arrives, then steadily declines to zero. Just when that inventory is depleted, a new order is received, thereby immediately sending the inventory level back to its maximum value (Q). This pattern continues throughout, with the inventory level fluctuating between Q and zero. To get a handle on the holding cost we are incurring, we can use the average inventory level throughout the year (which is Q/2). The cost of carrying those fluctuating inventory levels is equivalent to the cost that would be incurred if we had maintained that average inventory level continuously and steadily throughout the year. That cost would have been equal to the average inventory level times the cost to carry a unit in inventory for a year. In short, Total Annual Holding Cost = (Q/2)H Total Annual Cost The total annual relevant inventory cost would be the sum of the annual ordering cost and annual holding cost, or TC = (D/Q)S + (Q/2)H This is the annual inventory cost associated with any order size, Q.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 14 of 30

CLASSIC ECONOMIC ORDER QUANTITY (EOQ) MODEL

At this point we are not interested in any old Q value. We want to find the optimal Q (the EOQ, which is the order size that results in the lowest annual cost). This can be found using a little calculus (take a derivative of the total cost equation with respect to Q, set this equal to zero, then solve for Q). For those whose calculus is a little rusty, there is another option. The unique characteristics of the ordering cost line and the holding cost line on a graph are such that the optimal order size will occur where the annual ordering cost is equal to the annual holding cost. EOQ occurs when: (D/Q)S = (Q/2)H a little algebra clean-up on this equation yields the following: Q2 = (2DS)/H and finally ______ Q* = 2DS/H (Q* represents the optimal value for Q; this is what we call the EOQ)

Cost

Total Annual Cost

Annual Holding Cost

Annual Ordering Cost Economic Order Quantity (EOQ)

Lot Size (how much decision)

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 15 of 30

EOQ ILLUSTRATION
Given the following data for an inventory scenario whose characteristics fit the assumptions of the basic EOQ model: D = 15,000 units per year S = $3 per order H = $1 per unit per year LT = Replenishment lead time = 2 days Assume we have 300 operating days per year Find the following: 1. Average daily demand 2. EOQ 3. Number of orders placed per year 4. Total annual ordering cost 5. Total annual holding cost 6. Time between orders 7. Reorder point (in units) 8. Average inventory level Answers: 1. Average daily demand 15,000 units/yr 300 days/yr = 50 units per day ______ ______________ 2. EOQ = 2DS/H = (2)(15,000)(3)/(1) = 300 units/order 3. Number of orders placed per year D/Q = (15,000 units/yr)/(300 units/order) = 50 orders/yr 4. Total annual ordering cost (D/Q)(S) = [(15,000units/yr)/(300 units/order)]($3/order) = $150/yr 5. Total annual holding cost (Q/2)H = [(300 units/order/2)]($1/unit/yr) = $150/yr 6. Time between orders (Q/d) = (300 units/order)/(50 units/day) = 6 days/order [or, 300days/yr50 orders/yr = 6 days/order] 7. Reorder point (in units) ROP = (daily demand)(Lead time) = (50 units/day)(2 days) = 100 units 8. Average inventory level Q/2 = 300 units/2 = 150 units

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 16 of 30

OBSERVATIONS ABOUT OUR EOQ ILLUSTRATION

Results of computations EOQ = 300 units Number of orders placed per year = 50 Average inventory level = 150 units Annual ordering cost = $150 Annual holding cost = $150 Time between the placement of orders = 6 days Observation #1: Watch the inventory level instead of the calendar for when decision We discovered that our order quantity of 300 units would lead to a replenishment every 6 days. We projected that we would run out on days 6, 12, 18, 24, 30, 36, etc. With a 2 day lead time, we were smart enough to order 2 days in advance of when we would run out, which had us placing orders on days 4, 10, 16, 22, 28, 34, etc. We only have to watch the calendar to keep track of when those order instants arise so that we can place the orders. An alternative to watching the calendar would be to watch the inventory levels. Recall that the average daily demand for this item is 50 units per day. This means that at the moment we place an order, we have just enough inventory to cover the demand that will occur during the 2 day lead time. The demand during the 2 day lead time is 2 days x 50 units per day = 100 units. So, all we have to do is keep our eyes on our inventory level, and when it reaches 100 units, that is the signal that it is time to reorder. This level of inventory that triggers a reorder is called the reorder point (R). Inventory Level

EOQ = 300

ROP

100

Time

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 17 of 30

OBSERVATIONS ABOUT OUR EOQ ILLUSTRATION

Observation #2: Model is robust (insensitive to errors in estimates for input data) We estimated our holding cost to be $1/unit/yr when we made our EOQ calculation. Suppose this estimate was in error, and the actual holding cost that will be incurred is $2/unit/yr (an error of 100%!). If we had been aware of this true holding cost, and had used $2/unit/yr in our EQQ calculation, we would have determined the EOQ to be 212 units, and the ordering cost and holding cost would have each been $212, for a total annual cost of $424 (you can practice the application of the model to confirm these numbers on your own). But, unfortunately we were not aware that the holding cost would be $2/unit/yr, so we made our EOQ calculation using the incorrect $1/unit/yr. That calculation had us ordering 300 units each time we placed an order. With this order size, the true cost we will incur is as follows: Ordering cost: (D/Q)S = (15,000 units/yr/300 units/order)($3/order) = $150 Holding cost: (Q/2)H = (300 units/order/2)($2/unit/yr) = $300 Total annual cost = $150 + $300 = $450 In summary, we could have been incurring an annual cost of $424 if we had better information about the holding cost, and were ordering the correct EOQ of 212 units. But, we used the wrong holding cost in our model (we were off by 100%), ended up ordering 300 units every time we ordered, and incurred an annual cost of $450. Notice that the cost we are incurring ($450) is a little more than 6% higher than the absolute minimum cost ($424) that we might have incurred. Not bad. We made a 100% error on the input side, but our results are only about 6% worse than they could have been. That is because the total cost line on our cost graph is relatively flat in the vicinity of the EOQ. You can drift to the right or left of the optimal order size and find that the resulting cost doesnt rise substantially. This is what is meant by the model being robust (or insensitive to errors).

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 18 of 30

IMPACT OF CHANGING ASSUMPTIONS ON MODEL DEVELOPMENT

Our focus has been on the basic EOQ model. That is just one of dozens of inventory models for independent demand items. The basic EOQ model was derived from a set of underlying assumptions. If any of those assumptions do not fit a particular situation, then one must turn to a different model. Each of those available models is predicated upon a different set of underlying assumptions. Some of the more popular ones (and ones described in the textbook) are summarized below. We will not be expected to be knowledgeable about or work with any of these model extensions. Economic Production Quantity (EPQ) Model: When replenishment items come from inside sources, the entire batch is usually not received all at once (instantaneous replenishment), but instead is gradually received as a production batch is run (continuous replenishment). The pattern of inventory level fluctuations over time changes, resulting in a slightly different quantitative model for the optimal lot size. Quantity Discount Model: When the supplier is willing to offer a lower price if large quantities of an item are ordered, the total annual purchase cost line will no longer be horizontal, but will instead have step decreases in it. This will lead to a total cost curve that has breaks in its continuity (step changes) resulting in a slightly different model for determining the optimal order size. Controlled Backorder Model (not mentioned in the book): In some instances it might be beneficial to have shortages. If the backorder cost of a shortage is not very high, but the cost of carrying inventory is relative high, it may be more cost effective to incur some back orders on each order cycle (the saw tooth graph dips below the horizontal axis on each order cycle). This means that there will be less inventory being carried on average (resulting in lower holding costs) and some shortages that will incur some cost. How low below the horizontal axis this graph dips is a function of the relative values of the cost of holding inventory and the cost of incurring a shortage. Single-Period Inventory Model: Sometimes a unique situation that arises is one in which there will be demand for an item in only one period, so the challenge is to determine the order size (stock size) that will best accommodate the anticipated (and uncertain) demand. Any items stocked in excess of demand will be scrapped. Any demand in excess of what has been stocked will represent a missed opportunity for more profit. (This problem is sometimes referred to as the newsboy problem, or the Christmas tree problem.) These are but a few of the many variations to the basic EOQ model that are in existence. They all are designed to provide optimal answers to the how much and when questions. Choice of a model should be dictated by the characteristics of the inventory situation that you are facing.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 19 of 30

DEMAND VARIABILITY AND UNCERTAINTY

The basic EOQ model assumes that demand rate is constant and predictable. As a result we always knew when we were going to run out of inventory, so we could always reorder in a timely fashion so that the new replenishment order would be received just when we ran out of inventory. In reality demand rates are rarely constant and rarely completely predictable. It is more likely that demand rates will vary from day to day, and there will be uncertainty about what those demand rates will be at any one time. Consequently, there is a possibility that we may run out of inventory before a replenishment order arrives. To prevent a shortage situation organizations must rely on safety stock.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 20 of 30

ILLUSTRATION OF SAFETY STOCK DETERMINATION

Data: Average daily demand = 50 units per day Operating year contains 300 days of operation (D = 15,000 units per year) Ordering cost S = $3 per order Holding cost H = $1 per unit per year Lead time = 1 day Computations: EOQ (from EOQ formula) = 300 units per order Resulting number of orders per year = 50 orders per year Reorder point = 50 units (the average number of units demanded during the 1 day lead time) Additional Data: Demand is not always a constant 50 units per day. There is variability in daily demand according to the following table of demands and probabilities: Daily Demand Probability Cumulative Probability
Inventory Level

10 .01 .01

20 .04 .05

30 .05 .10

40 .2 .30

50 .4 .70

60 .2 .90

70 .05 .95

80 .04 .99

90 .01 1.00

300

300

Reorder Point, 50

Time
LT LT

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 21 of 30

The graph above suggests that if you waited until you had 50 units left in inventory before placing an order for 300 more units, you would be O.K. if the demand during the 1 day lead time was 10, 20, 30, 40, or 50. However, if the demand during the 1 day lead time was 60, 70, 80, or 90 you would have had a shortage. The size of the shortage would depend upon how many units were demanded during the lead time, but the maximum possible shortage would have been 40 units (if demand was the largest possible value of 90). You can prevent shortages by providing safety stock when there is uncertainty in demand. (Safety stock can be viewed as a cushion placed at the bottom of the saw tooth graph of inventory fluctuations over time.) If you wanted to guarantee that you would never have a shortage in this situation, you would need 40 units of safety stock at the bottom of the graph to "dip into" if demand spiked to higher than average values. But, adding 40 units of safety stock really means that you have elevated your reorder point. You are not waiting until there are only 50 units in inventory to place your order. You are ordering when there are 90 units in inventory. And, of course, 90 units are sufficient to cover the worst case scenario for this problem. The graph below illustrates the impact of 40 units of safety stock maintained in the system.
Inventory Level

300

300

Original Reorder Point, 50

Time Safety Stock, 40

LT LT

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 22 of 30

HOW MUCH SAFETY STOCK IS APPROPRIATE?

Service level: The probability that demand during lead time will not exceed the inventory on hand when the order is placed. In the previous illustration, it was suggested that you might provide 40 units of safety stock. If you had done so, you would never experience a shortage. You would have achieved a service level of 100%. This might not be a desirable solution for this problem. We are carrying a relatively high amount of safety stock, and there is a very low probability that lead time demand will actually go as high as 90 units (only a 1% chance). If you had chosen to carry only 30 units of safety stock (order when inventory drops to 80 units), you will be fine if lead time demand is anything up to and including 80 units. If lead time demand turns out to be 90 (there is a 1% chance of that), you will come up 10 units short. But, since you had enough inventory to cover 99% of the demands that might have occurred, you achieved a 99% service level. Many people might opt for this policy, for it will reduce the average annual level of inventory carried (i.e., reduce holding costs) and run only a slight risk of incurring a shortage cost. Others might be even more aggressive, and opt for an even lower service level. We could have achieved a 95% service level with a reorder point of 70 (only 20 units of safety stock). We've lowered our inventory holding costs even further, but exposed ourselves to even more shortage cost risk.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 23 of 30

INVENTORY MONITORING APPROACHES

Fixed Quantity, or Q System: This approach maintains a constant order size, but allows the time between the placement of orders to vary. This method of monitoring inventory is sometimes referred to as a perpetual review method, a continuous review system, a reorder point system, and a two-bin system. The inventory level is continuously (perpetually) monitored, and when the inventory drops to the reorder point level, a replenishment order is placed. The size of the order is constant (fixed quantity, typically the calculated economic order quantity for the item). Because demand continues to occur while we are waiting for the replenishment order to arrive (i.e., demand continues to occur during the lead time), the inventory level will generally be below the reorder point level when the replenishment arrives. This type of system provides closer control over inventory items since the inventory levels are under perpetual scrutiny.

Inventory Level

Order #2 Q Order #1 Q Order #3 Q

Reorder Point

LT

LT

LT

Time

How much decision: Order size is constant (fixed). When decision: Time between the placement of orders can vary.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 24 of 30

Periodic Review Systems: There are two, Fixed Period System (described in the textbook), and a hybrid system (described below but not in the textbook) Fixed Period System: This approach maintains a constant time between the placement of orders, but allows the order size to vary. This method of monitoring inventory is sometimes referred to as a fixed interval system. It only requires that inventory levels be checked at fixed periods of time. The amount that is ordered at a particular time point is the difference between the current inventory level and a predetermined maximum inventory level (also called an order up to level, or a target level). If demand has been low during the prior time interval, inventory levels will be relatively high when the review time occurs, and the amount to be ordered will be relatively low. If demand has been high during the prior time interval, inventory levels will have been depleted to low levels when the review time occurs, and the amount to be ordered will be higher. Since demand continues to occur during the lead time, inventory levels will increase when the replenishment order arrives, but not all the way up to the maximum (i.e., target) level.

Inventory Level Maximum (or order up to or target) inventory level

Order #2 Order #3

Order #1

Q2

Q3

Q1

Review time, T2 Review time, T3 Review time, T1

LT LT LT

T1

T2

T3

Time

How much decision: Order size can vary. When decision: Time between the placement of orders is constant (fixed).

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 25 of 30

Hybrid System: This approach allows both the order size and the time between the placement of orders to vary. This method of monitoring inventory is sometimes referred to as an optional replenishment system, or a min-max system. It is a hybrid system because it combines elements of both the fixed quantity system and the fixed period system. It is similar to the periodic review system in that it only checks inventory levels at fixed intervals of time, and it has a maximum inventory level (or order up to or target level). However, when one of those review periods arises the system does not automatically place an order. An order is only placed if the size of the order would be sufficient to warrant placing the order. This determination is made by incorporating the reorder point concept from the continuous review system. At the review period the inventory level on hand is compared to a minimum level for the item. If inventory has not fallen below this minimum level, no order is placed. However, if the inventory level has dropped below this minimum level, an order is placed. The size of the order is the difference between the inventory on hand and the maximum inventory level. Since demand continues to occur during the lead time, inventory levels will increase when the replenishment order arrives, but not all the way up to the maximum (i.e., target) level.

Inventory Level Maximum (or order up to or target) inventory level

Order #2

Order #1

Q2 Q1

Minimum Level
First order is placed at T1 since the inventory has fallen blow the minimum No order at T2 since inventory is not below the minimum Second order is placed at T3 since the inventory has fallen blow the minimum

LT

LT

T1

T2

T3

Time

How much decision: Order size can vary. When decision: Time between the placement of orders can vary.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 26 of 30

POSITIVES AND NEGATIVES OF INVENTORY MONITORING APPROACHES

Approach Fixed Quantity

Advantages (Positive aspects) - provides tighter control over inventory items - less safety stock needed - joint shipping advantage with multiple items from same source - does not require constant monitoring - joint shipping advantage with multiple items from same source - does not require constant monitoring - no small nuisance orders

Fixed Period

Hybrid

Disadvantages (Negative aspects) - requires constant monitoring (constant scrutiny) - problems with multiple items from same source (many items arrive in separate shipments) - requires more safety stock - occasional small nuisance orders may result - provides looser control over inventory items - requires more safety stock - provides looser control over inventory items

COMPARISON OF INVENTORY MANAGEMENT SYSTEMS

System Fixed Quantity Fixed Period Hybrid

Order Size Constant (fixed) Varies Varies

Time Between Orders Varies Constant (fixed) Varies

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 27 of 30

ABC CLASSIFICATION OF ITEMS

It is not unusual for organizations to maintain many items in inventory (hundreds or even thousands). Each of these items needs to be controlled. An important question is How much scrutiny does each item deserve? Some of these items may have a high annual investment, and logic would suggest that these items deserve very close scrutiny. On the other hand, some items may have a low annual investment (these are often referred to as nuts and bolts items), and they probably do not need as much attention. ABC analysis provides a mechanism to separate the "important few" from the "trivial many" so that the appropriate level of control can be assigned to each item. ABC analysis assigns all inventory items to one of these three classifications: A items need the tightest degree of control, while C items do not need very close scrutiny. The general graphical display for an ABC classification appears as follows: Cumulative % of Value 100%

A Items

B Items

C Items

Cumulative % of Items

100%

This type of diagram is referred to as a Pareto graph, and is relevant to a variety of situations.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 28 of 30

ILLUSTRATION OF ABC ANALYSIS

The following table displays an organizations inventory items, their value per unit, and their annual usage. (Note: To keep things manageable on the page, this illustration is greatly scaled down from reality. Most organizations would be dealing with considerably more than the ten inventory items displayed below.) Value Per Unit $13 $5 $6 $3 $5 $8 $10 $1 $5 $4 Total Rearrange items in decreasing order of annual dollar usage: Item Number 6 7 1 2 4 5 9 3 10 8 Total Annual $ Usage $400,000 $300,000 $130,000 $70,000 $30,000 $25,000 $20,000 $12,000 $8,000 $5,000 $1,000,000 % of Line Items 10% 10% 10% 10% 10% 10% 10% 10% 10% 10% Cumulative % of Items 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% % of Value 40 30 13 7 3 2.5 2 1.2 .8 .5 Inventory Item Number 1 2 3 4 5 6 7 8 9 10 Annual Usage 10,000 14,000 2,000 10,000 5,000 50,000 30,000 5,000 4,000 2,000 Annual Dollar Usage 130,000 70,000 12,000 30,000 25,000 400,000 300,000 5,000 20,000 8,000 $1,000,000 Cumulative % of Value 40 70 83 90 93 95.5 97.5 98.7 99.5 100 ABC Class* A A B B C C C C C C

*Note: When classifying the items as A, B, or C items, it can be somewhat subjective as to where the lines are drawn. With the unrealistically small demonstration above, the first 20% of the inventory items constitute 70% of the inventory value, so these items (Items 6 and 7) will be designated as A items. On the other extreme, 60% of the items constitute only 10% of the inventory value, so these items (Items 4, 5, 9, 3, 10, and 8) will be designated as C items. In the middle, 20% of the items constitute 20 % of the inventory value, so these items (Items 1 and 2) will be designated as a B item.

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 29 of 30

DEVELOPMENT OF THE ABC INVENTORY PARETO GRAPH

Cumulative percentages extracted from previous table (and rotated 90 degrees) Items (rearranged order) Cumulative % of items Cumulative % of value
1st 1 10 40 1st 2 20 70 1st 3 30 82 1st 4 40 90 1st 5 50 93 1st 6 60 95.5 1st 7 70 97.5 1st 8 80 98.7 1st 9 90 99.5 All 10 100 100

MAN 3520 Fall 2012 CP3 Independent Demand Inventory: Page 30 of 30

ALTERNATE REPRESENTATION OF OUR ABC ANALYSIS

The data reflected in the Pareto graph could also be displayed in a bar chart, as illustrated in the textbook. The following is such a representation for our simple ABC illustration.

A items

B items C items

10 80

20 90

30 100

40

50

60

70