yor pee FIXED-ORDER QUANTITY MODELS CEO Fed-onder uanity models tempt to determine te specif poi, Rat which a order wil he paced andthe ize of that ode, QT ode point, salvaje speed tombe of unis, An oder of size Qi placed when the inventory ave (cumeny stack td on order) reaches the pit Inventory postions efind as the ow hand pls on oder minus backordered quant. The voution fou fxed-order quantity model may stipulate ‘omehing Iie is When be inventory poston drops to 3, place an cde for 37 ore nits The simplest models inthis categrycccur when all aspects ofthe sitvaton are Kaew wih enti the anna demand fer «produc is 1,000 units is precisely 1,000 not 1,000 pls or mts 10 percent. The sme tre for setup costs and boli cous. Albough the assumption of complete cersiny israel vali, provide «good busi fr our coverage ieee Exhibit 153 adhe dscusion about driving te optimal rer quantity are based onthe folowing characteristic ofthe model, Tes sumptions ae aneaiic, ut hey represent {tating pot lbw os aes ag ‘+ Demand forthe product is constant and uniform throughout the period, + Lead time (time from ordering to receipt is constant. + Price per unit of product is constant, + Inventory holding cost is based on average inventory. + Ordering o setup costs are constant +All demands fr the product willbe said. (No backorders are allowed.) ‘The “sawtooth effect” relating Q and R in Exhibit 15.3 shows that when the inventory position drops to point, a reorders placed. This order i rceived a the end of time period 1, which does not vary inthis model In constructing any inventory model the first step isto develop a functional relationship between the variables of interest and the measure of effectiveness In tis case, because we are concerned with cos, the following equation pertains Tol Annual ‘Annual Annual annual cost ~ purchase cost * ordering cost * holding cost De 183] te=0c+2s42u 3 ors lap 15 97 Inventory position Turia £09exhibit 15.4 "Annual Product Costs, Based on Size ofthe Order i lilt 8 eign em ftom Bs oteng om Ona rier uti sie (0) Demand (annual) Cost per unit (@ = Quantity 1 be ordered (the optimal amount is termed the economic order quantity EOQ—0" Qn) Setup cost or cost of placing an order Reorder point Lead time H = Annual holding and storage cost per unit of average inventory (often holding cost is taken asa percentage ofthe cost of the item, such as 1 = iC, where is the pereent carying cost) (On the sight side of the equation, DC isthe annual purchase cost for the units, (D/Q)S is the annual ordering cost (the actual number of orders placed, D/@. times the cost of each fonder, $), and (Q/2)H is the annual holding cost (he average inventory, Q/2, times the cost pet unt for holding and storage, H). These cos relationships are graphed in Exhibit 15.4 “The second step in model development is ofind that order quantity Qu, at which total cost is a minimum. In Exhibit 154, the total cos is minimal atthe point where the slope ofthe ‘curve is zero. Using calculus, we take the derivative of total cost with respect to Q and set this equal to zero, For the basic model considered here, the calculations are Dae Te=De+os+ SH mo le #81Iyvenrony Gornon chapter 15 Because this simple model assumes constant demand and lead time, no safety stock is necessary, andthe reorder point, R, is simply psa) R where J = Average daily demand (constant) 1L = Lead time in days (constant) EXAMPLE 52 Economic Order Quanity and Reorder Point on Find te economic order quantity and the reorder pin, given a Anna demand (2) = 1.00 ois fice ‘Average iy demand (2) = 100/365 inventory ‘Onlering cost (S) = 85 por ower Holding cost) = 1.25 per unt per year Lead ime (2) = 5 days Cost perunit (C) = $12.50, ‘What quantity shouldbe ordered? SOLUTION “The optimal order quantity is ees 20-0905 06 89.4 units “The reorder point is ing tothe nearest uni, the inventory policy is as fallows: When the inventory positon ops to 14, place an onerfor 89 more. ‘The toa annual eos wil be ‘ 9 Pay =sr261181 [Note that inthis example, the purchase cost ofthe units was not required to determine the order ‘quantity andthe reorder point Because the cost was constant anid unrelated to onder size. ESTABLISHING SAFETY STOCK LEVELS ‘The previous model assumed that demand was constant and known. In the majority of cases, though, demand isnot constant but varies from day today. Safety stock must therefore bbe maintained to provide some level of protection againststockouts. Safety stockcanbedefined Safety stock asthe amount of inventory carried in addition tothe expected demand, laa normal distribution, 599600 section 4 Puaysine anb CostnonLiNe i: SuPALy ‘this would be the mean, For example, ifuraverage monthly demand s 100 units and we expect next month tobe the same, if we eary 120 uit, then we have 20 units of safety stock. Safety stock ean be determined based on many different criteria, A common approach is fora company to simply stat tata certain number of weeks of supply be kept in safety stock. kis better, though, to use an approach that captures the variability in demand. For example, an objective may be something like “set the safety stock evel so that there will only be a $ percent chance of stocking out if demand exceeds 300 units.” We call this approach to setting safety sock the probability approach. The Probability Approach Using the probability criterion to determine safety ‘stock is pretty simple, With the models described inthis chapter, we assumne thatthe demand ‘over a period of time is normally distributed with a mean and a standard deviation. Again, remember that this approach considers only the probability of runing out of stock, mot how ‘many units we are short, To determine the probability of stocking out over the time period, ‘we can simply plot « normal distribution for the expected demand and note where the amount wwe have on hand lies on the curve. Ler’s take a few simple examples to illustrate this. Say we expect demand to be 100 units ‘over the next month, and we know thatthe standard deviation is 20 units. If we go into the ‘month with ust 100 units, we know that our probability of stocking outs 50 percent, Half of the months we would expect demand to be greater than 100 units; half of the months we ‘would expect it tobe less than 100 units. Taking this further, if we ordered a month's worth of inventory of 100 units at atime and received it atthe beginning ofthe month over the long, ‘un we would expect to run out of inventory in six months of the year. If running out this often was not acceptable, we would want to carry extra inventory 10 reduce this risk of stocking out. One idea might be to carry an extra 20 units of inventony forthe item, In this ease, we would still order a month’s worth of inventory at atime, but wwe would schedule delivery to arrive when we still have 20 units remaining in inventory. ‘This would give us that litle cushion of safety stock to reduce the probability of stocking. out. Ifthe standard deviation associated with our demand was 20 units, we would then be carrying one standard deviation worth of safety stock. Looking at the Cumulative Standard [Normal Distribution (Appendix E), and moving one standard deviation to the right of the ‘mean, gives a probability of .8413. So approximately 84 percent of the time we would not expect to stock out, and 16 percent of the time we would. Now if we order every month, ‘we would expect {0 stock out approximately two months per year (16 x 12 = 1:92) those using Excel, given az value, the probability can be obtained with the NORMSDI function. is common for companies using this approach to set the probability of not stocking 495 pent. This means eu cay shout 1.64 standard deviations of sly slock 33 units (1.64 x 20 = 32.8) fOr our example. Once again, keep in mind that this does ‘ot mean that we would order 33 units extra each month. Rather, it means that we would sil order a month's worth each time, but we would schedule the receipt so that we could expect to have 33 units in inventory when the order arrives. In this ease, we would expect to stock ‘ut approximately 6 month pee year, or that stockouts would occur in 1 of every 20 months. eRres Fixep-ORDER QUANTITY MODEL WITH Sarety STOCK [A fixed-order quantity system perpetually monitors the inventory level and places a new ‘order when stock reaches some level, R. The danger of stockout inthis model occurs only. “ding the lead time, between the time an order i placed and the time itis received. As shown. in Exhibit 15.5, an onder is placed when the inventory position drops to the reorder point, R- Daring this lead time Z, range of demands is possible. This range is determined either from 1m analysis of past demand data or from an estimate (if past data are not availabe). “The amount of safety stock depends on the service level desired, as previously discussed. “The quantity to be ordered, Q, is calculated in the usual way considering the demand, short= age cost, ordering cost, holding cost, and sofort. A fixed-order quantity model can be used BORSEFRO z