Inventory Planning
&
Control
Probabilistic Demand
• Safety stock determines the chance of a stockout during lead time
• The complement of this chance is called the service level
• Service level is defined as the probability of not incurring a stockout
during any one lead time
• The higher the probability inventory will be on hand, the more likely
customer demand will be met.
• Service level of 90% means there is a .90 probability that demand will be
met during lead time and .10 probability of a stockout.
Probabilistic Demand
Probabilistic Demand
Use prescribed service levels to set safety stock when the cost of stockouts
cannot be determined
ROP = demand during lead time + ZsdLT
where Z = Number of standard deviations
sdLT = Standard deviation of demand during
lead time
Probabilistic Demand
Probabilistic Demand
Probability of Risk of a stockout
no stockout (5% of area of
95% of the time normal curve)
Mean ROP = ? kits Quantity
demand
350
Safety
stock
0 z
Number of
standard deviations
Probabilistic Demand
m = Average demand = 350 kits
sdLT = Standard deviation of
demand during lead time = 10 kits
Stock-out policy = 5% (service level = 95%)
For an area under the curve of 95%, the Z = 1.645
Safety stock = ZsdLT = 1.645(10) = 16.5 kits
Reorder point= Expected demand during lead time + Safety stock
= 350 kits + 16.5 kits of safety stock
= 366.5 or 367 kits
Other Probabilistic Models
When data on demand during lead time is not available, there are other
models available
1. When demand is variable and lead time is constant
2. When lead time is variable and demand is constant
3. When both demand and lead time are variable
Other Probabilistic Models
Other Probabilistic Models
Demand is variable and lead time is constant
ROP = (Average daily demand x Lead time in days) + ZsdLT
where sdLT = sd Lead time
sd = Standard deviation of demand per day
Other Probabilistic Models
Other Probabilistic Models
Lead time is variable and demand is constant
ROP = (Daily demand x Average lead time in days) + Z x (Daily demand) x sLT
where sLT = Standard deviation of lead time in days
Other Probabilistic Models
Other Probabilistic Models
Both demand and lead time are variable
ROP = (Average daily demand x Average lead time) + ZsdLT
where sd = Standard deviation of demand per day
sLT = Standard deviation of lead time in days
sdLT = (Average lead time x sd2)
+ (Average daily demand)2s2LT
Other Probabilistic Models: Generic Variance
Calculation
Other Probabilistic Models
Because daily demands are independent and identically distributed
E[X] = E[L] * E[D] =LD
• However, variable lead times change the variance of demand during replenishment lead time.
• Using the standard formula for sums of independent, identically distributed RVs,
• Var (X) = E[L] Var(D) +E[D]2 +Var [L] = LσD2 + D2σL2
• Standard Deviation= 𝑉𝑎𝑟 [𝑋]
Single-Period Model
• Only one order is placed for a product
• Units have little or no value at the end of the sales period
• The single-period inventory model is second only to the economic
order quantity in its wide-spread use and influence.
• Also referred to as the Newsvendor or (the Newsboy model)
Single-Period Model vs EOQ
• EOQ assumes uniform and deterministic demand, the single-period model allows demand to be
variable and stochastic (random).
• EOQ assumes a steady state condition (stable demand with essentially an infinite time horizon),
the single-period model assumes a single period of time.
• All inventories must be ordered prior to the start of the time period and it cannot be replenished
during the time period. Any inventory left over at the end of the time period is scrapped and cannot
be used at a later time. If there is extra demand that is not satisfied during the period, it too is lost.
• For EOQ we are minimizing the expected costs while for the single period model we are actually
maximizing the expected profitability.
Single-Period Model : Costs Associated
Only one order is placed for a product
Units have little or no value at the end of the sales period
Cs = Cost of shortage = Sales price/unit – Cost/unit
Co = Cost of overage = Cost/unit – Salvage value
If we consider the continuous distribution of demand
Expected Excess Cost of Qth unit ordered = Co* P [X≤Q]
Expected Shortage cost of the Qth unit ordered =Cs * (1-P[X ≤Q])
Single-Period Model : Costs Associated
If E[Excess Cost] < E[Shortage Cost] then increase Q
We are at Q* when E[Shortage Cost] = E[Excess Cost]
Cs
Critical Ratio, Service level =
Cs + Co
Expected Excess Cost of Qth unit ordered = Co* P [X≤Q]
Expected Shortage cost of the Qth unit ordered =Cs * (1-P[X ≤Q])
Single-Period Model
The newsstand near me usually sells 120 copies of the Times of India
each day. The owner of this stand believes the sale of the newspaper is
normally distributed with standard deviation of 15 papers. He pays 70
paisa for each paper which sells for Rs 1.25. The paper gives him a 30
paisa credit for each unsold paper. He wants to determine how many
papers he should order each day and the stock out risk for that
quantity.
Single-Period Model
Average demand = µ = 120 papers/day
Standard deviation = s = 15 papers
Cs = cost of shortage = ₹1.25 – 70p = 55p
Co = cost of overage = 70p – 30p = 40p
Cs
Service level =
Cs + Co
.55 Service
= level
.55 + .40 57.9%
.55
= = .579 µ = 120
.95
Optimal stocking level
For an area under the curve of .579, the Z = .195
Fixed-Period (P) Systems
• Fixed-quantity models require continuous monitoring using perpetual
inventory systems
• In fixed-period systems orders placed at the end of a fixed period
• Periodic review, P system
• Inventory is only counted at each review period
• May be scheduled at convenient times
• Appropriate in routine situations
• May result in stockouts between periods
• May require increased safety stock
Fixed-Period (P) Systems
Figure 12.9
Target quantity (T)
Q4
On-hand inventory Q2
Q1 P
Q3
Time
