Inventory Management and Risk Pooling

Question 1

Why hold inventory?

Answer 1

1. Hedge against uncertainty in supply & demand
2. Economies of scale
3. Hedge against lead time
4. Capacity limitations

Question 2

Order costs

Answer 2

Product cost, transportation cost (straightforward computation)

Question 3

Holding costs

Answer 3

Capital tied up, physical costs: warehouse space, storage tax, insurance, breakage, spoilage

Question 4

Other inventory costs (other than order and holding costs)

Answer 4

Component devaluation costs (life cycle dependent)
Price protection costs (supply contract dependent)
Product return costs (operational cost)
Obsolescence cost (FG inventory + components in pipeline + probable discount)
Out of stock costs (difficult)

Question 5

Tendency to overstock (diagram)

Answer 5

Cost of stock outs high, cost of obsolescence low

Question 6

Tendency to understock (diagram)

Answer 6

Cost of stock outs low, cost of obsolescence high

Question 7

Balancing act

Answer 7

Cost of stock outs high, cost of obsolescence high

Question 8

Forecasting methods

Answer 8

Quantitative - moving average, exponential smoothing

Qualitative methods

Question 9

Principles of forecasts

Answer 9

1. Forecasts always wrong
2. Longer the horizon, the worse the forecast
3. Aggregate forecasts always more accurate

Question 10

Three types of inventory

Answer 10

Raw-materials
WIP
Finished goods

Question 11

Four main methods to keep track of inventory and differences

Answer 11

1. General of average rules
2. ABC analysis
3. MRP or APS systems
4. Multi echelon inventory system
   MRP APS - demand uncertainty
   Multi echelon - capacity considerations

Question 12

ABC analysis

Answer 12

Split products in to A, B, C
A - 80% revenue 20% inventory space
B - 15% revenue 30% inventory space
C - 5% revenue 50% inventory space

Question 13

Cycle service level

Answer 13

Fraction of replenishment cycles

with no stock out

Question 14

Fill rate

Answer 14

Fraction of demand satisfied from stock on hand

Question 15

Prediction Methods

Answer 15

Judgement, market research, time series, causal methods

Question 16

Judgement methods

Answer 16

Opinion of experts

Delphi method - panel of experts share opinions independently over several rounds

Question 17

Marker research methods

Answer 17

Qualitative forms - Base predictions solely on behaviour of consumers

Question 18

Time series

Answer 18

Data from the past used to predict the future, quantitative method. Moving average or exponential smoothing

Question 19

Causal methods

Answer 19

Predictions based on basis of sales data and how it differs from predictions. Relies on external factors

Question 20

Economic lot size model characteristics

Answer 20

Demand is known and constant
Order quantity fixed at Q per order
Balance fixed cost K and inventory holding cost h
No lead time
No discounts
No stock outs

Question 21

Single period model

Answer 21

Data from the past, uses this data to make predictions and future scenarios. Only one order moment with no backorders

Question 22

Multiperiod model

Answer 22

Several orders with order thresholds and order sizes

R, Q), (s, S

Question 23

(R, Q) policy

Answer 23

Fixed order quantity Q, once reorder point triggered R.

Reorder point = avg demand during lead time + safety stock

Question 24

Periodic checks without fixed costs (S-1, S)

Answer 24

S represents order up to point. Inventory placed each check - order up to certain amount. r is the review period time between two periodic checks

Question 25

Periodic checks (s, S)

Answer 25

Min-max model. s = reorder point. S = order up to level

Question 26

Risk pooling (pros cons)

Answer 26

Pros: Demand variability reduced when aggregated across locations. Reduction in variability allows for decrease in SS reducing avg inventory, while maintaining same service level
Cons: increase in response time and possible increase in transportation costs
Benefits pronounced where supplied to regions where demand is negatively correlated

Question 27

Square root law (risk pooling)

Answer 27

If demand in k geographic locations is independent and about same size,
Then aggregation reduces safety inventory by about a factor of squareroot(k)

Question 28

Coefficient of variation

Answer 28

Ratio of STDEV to average demand - gives relative measure of variability relative to average demand.
The higher the coefficient of variation, the greater the benefit from risk pooling

Question 29

Centralised vs. Decentralised systems
SS
Service level
Overhead costs
Customer lead time
Transportation costs

Answer 29

SS - lower with centralisation
Service level - higher service level with same investment with centralisation
OH costs - higher in decentralised
Customer lead time - lower in decentralised
Transport costs - Not clear. Outbound vs inbound costs