Inventory Management Subject to Uncertain Demand
ISYE 3104 - Fall 2012 nar Keskinocak, 2007 nar Keskinocak, ı © Esma Gel, Gel, P © Esma
Inventory Control Subject to Uncertain Demand
In the presence of uncertain demand, the objective is to minimize the expected cost or to maximize the expected profit
Two types of inventory control models
Fixed time period - Periodic review
One period (Newsvendor model)
Multiple periods
Fixed order quantity - Continuous review
(Q,R) models
1 Types of Inventory Control Policies
Fixed order quantity policies
The order quantity is always the same but the time between the orders will vary depending on demand and the current inventory levels
Inventory levels are continuously monitored and an order is placed whenever the inventory level drops below a prespecified reorder point.
Continuous review policy
Types of Inventory Control Policies
Fixed time period policies
The time between orders is constant, but the quantity ordered each time varies with demand and the current level of inventory
Inventory is reviewed and replenished in given time intervals, such as a week or month (i.e., review period)
Depending on the current inventory level an order size is determined to (possibly) increase the inventory level up-to a prespecified level (i.e., order- up-to level)
Periodic review policy
2 Inventory Control - Demand
Variability Constant/Stationary Variable/Non-Stationary
Economic Order Aggregate Planning – Quantity (EOQ) – Planning for capacity Tradeoff between levels given a forecast fixed cost and holding Materials Requirements cost
Deterministic Planning (MRP)
Uncertainty Lot size/Reorder point (Q,R) or (s,S) models – Very difficult problem! Tradeoff between fixed cost, holding cost, and
Stochastic shortage cost
Newsvendor – single period
READ THE APPENDIX ON PROBABILITY REVIEW
3 Newsvendor Models
ISYE 3104 – Fall 2012 nar Keskinocak, 2007 nar Keskinocak, ı © Esma Gel, Gel, P © Esma
Example - INFORMS
INFORMS (The Institute for Operations Research and Management Science, www.informs.org) will hold its annual meeting in Washington D.C. in 2008. Six months before the meeting begins, INFORMS must decide how many rooms should be reserved at the conference hotel. At this time, rooms can be reserved at a cost of $50 per room. It is estimated that the demand for rooms is normally distributed with mean 5000 and standard deviation 2000. If the number of rooms required exceeds the number of rooms reserved, extra rooms will have to be found at neighboring hotels at a cost of $80 per room. The inconvenience of staying at another hotel is estimated at $10. How many rooms should be reserved to minimize the expected cost?
4 Example – Fashion Bags
The buyer for What-a-Markup Fashion Bags must decide on the quantity of a high-priced woman’s handbag to procure in Italy for the following Christmas Season. The unit cost of the handbag to the store is $28.50 and the handbag will sell for $150. A discount firm purchases any handbags not sold by the end of the season for $20. In addition, the store accountants estimate that there is cost of $0.40 for each dollar tied up in inventory at the end of the season (after all sales have been made).
Newsvendor model - Properties
One-time decision
Current decisions only impact the “next period” but not future periods
Retail: fashion/seasonal items
One-time events
5 Tradeoff in inventory decisions
Demand Supply Shortage Supply < Demand Lost sales / Lost profit
Excess inventory Supply > Demand Inventory cost
Supply Demand
Newsvendor model - Properties
One-time decision
Current decisions only impact the “next period” but not future periods
Retail: fashion/seasonal items
One-time events Relevant costs
Co: Unit cost of excess inventory (cost of overage) Cu: Unit cost of shortage (cost of underage) Demand D is a random variable with
Probability density function (pdf), f(x)
cumulative distribution function (cdf), F(x) Objective: Choose the ordering quantity Q (before you know the demand) to minimize
Total expected overage and underage costs
6 Newsvendor model – Finding the optimal order quantity
First, write the cost function:
G(Q, D) :total overage underage cost (if Q is ordered and demand is D)
cu (D Q) if D Q G(Q, D) : co (Q D) if D Q
G(Q) expected overage underage cost if Q is ordered G(Q) E[G(Q, D] G(Q, x) f (x)dx Critical Ratio 0 Q c c (Q D) f (x)dx c (D Q) f (x)dx F(Q) u o u 0 Q cu co
Derivation of the Critical Ratio - 1
G(Q) expected overage underage cost if Q is ordered Q G(Q) c (Q D) f (x)dx c (D Q) f (x)dx c G c G o u o 1 u 2 0 Q
To take the derivative of G(Q), use Leibniz's Rule d f2 ( y) f2 ( y) h(x, y) h(x, y)dx dx h( f (y), y). f ' (y) h( f (y), y). f ' (y) dy y 2 2 1 1 f1 ( y) f1 ( y)
In G1: y Q f2 (y) Q f1(y) 0 h(x,Q) (Q x) f (x)
7 Derivation of the Critical Ratio - 2 Q G (Q) (Q D) f (x)dx In G1: y Q f2 (y) Q f1(y) 0 1 0 h(x,Q) (Q x) f (x) To take the derivative of G(Q), use Leibniz's Rule d f2 ( y) f2 ( y) h(x, y) h(x, y)dx dx h( f (y), y). f ' (y) h( f (y), y). f ' (y) dy y 2 2 1 1 f1 ( y) f1 ( y)
h(x, y) [(Q D) f (x)] f (x) y Q ' h( f2 (y), y) h(Q,Q) (Q Q) f (x) 0 f2 (y) 1 ' h( f1(y), y) h(0,Q) (Q 0) f (x) Qf (x) f1 (y) 0 d Q d G (Q) f (x)dx Similarly, G (Q) f (x)dx 1 2 dQ 0 dQ Q
Derivation of the Critical Ratio - 3
d Q G(Q) c f (x)dx c f (x)dx c F(Q) c (1 F(Q)) o u o u dQ 0 Q
(co cu )F(Q) cu 0 c F(Q) u cu co
Recall: f (x)dx 1 0 Q f (x)dx F(Q) P(D Q) f (x)dx 1 F(Q) P(D Q) 0 Q
Also need to check if G(Q) is convex. Second derivative is (co+cu)f(Q)≥0
8 Example - INFORMS
D = number of rooms actually required
Q = number of rooms reserved
What are Co and Cu ?
Example - INFORMS
D = number of rooms actually required
Q = number of rooms reserved
DQ: cost = 50Q DQ: cost = 50Q+80(D-Q)+10(D-Q) =90D-40Q
Cost of Cost of Cost of reserving reserving inconvenience extra rooms in other hotels
9 Example - INFORMS
D = number of rooms actually required
Q = number of rooms reserved
Co DQ: cost = 50Q Cu DQ: cost = 50Q+80(D-Q)+10(D-Q)=90D-40Q
Cost of Cost of Cost of reserving reserving inconvenience extra rooms in other hotels
cu 40 4 F(Q) What is Q? cu co 40 50 9
Example - INFORMS
Recall: D Normal(5000,2000)
We have lookup tables for Standard Normal distribution ( =0, =1) Convert to Standard Normal!
4 F(Q) P(D Q) 0.444 9 D Q P(D Q) P(D Q ) P P(Z z) (z) Standard normal Zz
Find z from the lookup table Q = z +
10 Example - INFORMS
Recall: D Normal(5000,2000)
We have lookup tables for Standard Normal distribution ( =0, =1) Convert to Standard Normal!
4 F(Q) P(D Q) 0.444 9 D Q P(D Q) P(D Q ) P P(Z z) (z) Safety Standard normal Zzstock Expected demand
Find z from the lookup table Q = z +
Standard Normal Distribution - 1
z 2 e 2 Density function (z) 2 Symmetric around the mean Area under the entire curve 1 P(Z )
11 Standard Normal Distribution - 2
z 2 e 2 Density function (z) We will use ɸ(z) and F(z) 2 interchangeably Symmetric around the mean Area under the entire curve 1 P(Z ) F(z)=P(Z≤z)
z
Standard Normal Distribution - 2
z 2 e 2 Density function (z) We will use ɸ(z) and F(z) 2 interchangeably Symmetric around the mean Area under the entire curve 1 P(Z ) F(z)=P(Z≤z)
F(-z)=P(Z≤-z)=1-F(z)
-z z
12 Example - INFORMS
P(Z z) (z) Area under the curve 0.444 z -0.14
Q* z 2000(0.14) 5000 Q* 4720
z = - 0.14
Example - INFORMS
Why is Q*<5000, i.e., less than the expected demand? Expected total cost Expected cost of underage
Expected cost of overage
Q*=4720
13 Example - INFORMS
What if co=150? Expected c 40 total cost F(Q) u cu co 40 150 0.21 z 0.805
Expected Q* z Expected cost of (2000)(0.805) 5000 cost of overage 3390 underage
Q*=3390
Example - INFORMS
What if co=150? Expected c 40 total cost F(Q) u c c 40 150 u o C Q* 0.21 z 0.805 o Expected Cu Q* Q* z Expected cost of (2000)(0.805) 5000 cost of overage 3390 underage
Q*=3390
14 When do we have Q*=Expected demand? Q* z If Q* then z 0 !!!
P(Z z) (z) Area under the curve 0.5 z 0
For Q* we need
cu F(Q) 0.5, i.e., cu co!!! cu co
z = 0
Optimal quantity versus expected demand
cu cu co F(Q) 0.5 z 0 Q* cu co If shortages are more costly, order more than expected demand
cu cu co F(Q) 0.5 z 0 Q* cu co If excess inventory is more costly order less than expected demand
cu cu co F(Q) 0.5 z 0 Q* cu co If shortages and excess inventory cost the same, order expected demand
Note: Assuming the demand distribution is symmetric around its mean
15 The impact of standard deviation
What happens to the optimal quantity as the standard deviation increases?
The impact of standard deviation
What happens to the optimal quantity as the standard deviation increases? (Assuming the demand distribution is symmetric around its mean)
cu co z 0 Q* z increases in If shortages are more costly, Q* increases in std. dev.
cu co z 0 Q* z decreases in If excess inventory is more costly , Q* decreases in std. dev.
cu co z 0 Q* does not change in If shortages and excess inventory cost the same, order expected demand regardless of standard deviation
16 Example - INFORMS
cu=40 < co=50 z=-0.14 cu=70 > co=50 z=0.21
Q* Q*
Standard deviation Standard deviation
Example – Fashion Bags Input
Unit cost c = $28.50
Selling price p = $150
Salvage value s = $20
Cost of inventory = $0.40 for each dollar tied up in inventory at the end of the season
17 Example – Fashion Bags Input
Unit cost c = $28.50
Selling price p = $150
Salvage value s = $20
Cost of inventory = $0.40 for each dollar tied up in inventory at the end of the season
Computed input:
Holding cost per bag h =
cu =
co = c F(Q) u cu co
Example – Fashion Bags Input
Unit cost c = $28.50
Selling price p = $150
Salvage value s = $20
Cost of inventory = $0.40 for each dollar tied up in inventory at the end of the season
Computed input:
Holding cost per bag h = (0.40)(28.50) = $11.4
cu = p – c = $121.5
co = c – s + h = $19.9 c 121.5 F(Q) u 0.86 cu co 121.5 19.9
18 Example – Fashion Bags – Normally distributed demand
Demand ~ Normal( 150, 20)
F(Q) 0.86 z 1.08 Area under the curve= Critical ratio = 0.86
Q* z (20)(1.08) 150 Q*172
z = 1.08
Example – Fashion Bags – Uniformly distributed demand
Demand Uniform between 50 and 250 =150
Same expected demand as in Normal distribution
1 0.86 (Q*50) 200 Uniform density Area under the curve = Q* 222 Critical ratio = 0.86
1/200
Q 50 250
Q*= 222
19 Example – Fashion Bags
Even though both the Normal and the Uniform distributions have the same mean (=150), why did we get different quantities?
Normal distribution Q*=172
Uniform distribution Q*=222
Example – Fashion Bags
Even though both the Normal and the Uniform distributions have the same mean (=150), why did we get different quantities?
Normal distribution Q*=172
Uniform distribution Q*=222
Because of the variance (equivalently, standard deviation) and the shape of the distribution !!!
Normal =20
Uniform =57.7 Normal
Uniform
20