Inventory Optimization

Question 1

Key task of inventory optimization

Answer 1

Balance service level and inventory costs

Question 2

Implement’s 8 viewpoints

Answer 2

1) Breakdown inventory and levers for stock level
2) Top management must balance SL and cost
3) Transparent simulation
4) Understand variability
5) No size fits all
6) Normal distribution = fill rate
7) Lumpy, erratic, intermittent = special model
8) Standardized process for optimization

Question 3

Why is adherence to the parameters important?

Answer 3

If we create an optimal plan, then it is only optimal as long as it is carried out by the planner. If the planner decides to deviate from the plan due to missing trust, then the plan and process is no longer optimal

Question 4

What are the two reasons to carry stock?

Answer 4

Strategic reason:

• Product is critical to customers
• Product have high impact on customer perception

Supply chain reason:

• Is unit costs very low?
• Long lead time?

Question 5

What are the different types of inventory?

Answer 5

Safety: Buffer for serving unstable demand

Cycle: Normal inventory

Pipeline: Inventory in the pipeline due to lead times and internal transportation

Excess: Products not sold due to phase-outs, bias in forecasts, obselences or dead-stock.

Decoupling:

Anticipation: Used to handle uncertainty in the future from seasonality, disruptions or phase-in of products

Strategic: To mitigate risks of SC disruptions

Question 6

Describe the trade-off between product availability and costs?

Answer 6

High product availability or service levels means that many customers get their demand served from the inventory, which yields high revenues.

However, it requires high level of inventories, which will increase costs. Therefore, there must be a balance.

Question 7

Describe the differences between traditional and IBP approach to inventory

Answer 7

Traditional:

• All plan for themselves
• Target SLs are set on location level
• Suboptimizing in isolation

IBP:

• Viewing the SC as a whole
• E2E inventory policy
• Target SLs set on customers
• Focus on NWC

Question 8

Describe demand variability

Answer 8

Demand variability describes the uncertainty of the demand during the lead time. The safety stock much cover this, therefore:
Longer lead time = more variability = higher safety stock required

Question 9

Describe lead time variability

Answer 9

Lead time variability describes the uncertainty of the time it takes to replenish due to variability in production, quality or transportation from the supplier.
High lead time variability = higher safety stock

Question 10

Describe supplier variability

Answer 10

Supplier variability describes the variability in the suppliers ability to deliver certain order sizes. This might be due to component shortage, capacity issues, quality issues.
Higher variability = higher safety stock

Question 11

What are the forecast error measurements?

Answer 11

Mean absolute deviation percentage MADP

Root mean square error percentage RMSEP or CV(RMSE)

Question 12

How to measure demand variability?

Answer 12

Using Coefficient of variation of historical demand CV(D)

Question 13

When to use CV(RMSE) or CV(D)

Answer 13

We can only use CV(RSME) when we have a good forecast at hand. Otherwise, CV(D).

Question 14

Downside of CV(D)?

Answer 14

Does not take trends, seasonality, promotions into account.

Question 15

What is an important aspect when looking at demand variability?

Answer 15

Number of order lines per lead time

Question 16

If number of order lines per lead time is < 3, then what?

Answer 16

Cause for variability: Random, unpredictable demand

Use Poisson model for safety stock setting

Question 17

If number of order lines per lead time is < 8, then what?

Answer 17

Cause for variability: Demand is predictable

Focus on understanding high-impact materials

Question 18

If number of order lines per lead time is 3-8, then what?

Answer 18

Cause for variability: Order size variability

Use compound Poisson

Question 19

If number of order lines per lead time is > 8, then what?

Answer 19

Cause for variability: Seasonality, trend, promotion, step-changes

Use fill-rate model and reduce safety stock by improving forecast

Question 20

What drives demand variability?

Answer 20

Few order lines

Order size variability

Question 21

Describe the categorization matrix for demand

Answer 21

CV low + Few lines = Intermittent

CV low + many lines = Normal

CV high + many lines = Erratic

CV high + few lines = Mix

Question 22

What characterizes normal distributed demand?

Answer 22

Loew variance and frequent demand
No forecasting or inventory issues
Use norm. distr.

Question 23

What characterizes erratic demand?

Answer 23

high variance and frequent demand
Variance stems from unpredicted spikes, seasonality, or forecast bias.

Use norm. distr. with spike order procedures and reduce bias
Potential use gamma

Question 24

Intermittent demand

Answer 24

low variance and low frequency

Use norm. distr.

Question 25

Lumpy demand

Answer 25

high variance and infrequent demand

Difficult to manage products

Use compound poisson with spike handling procedures

Question 26

Describe why fill-rate method is better than service level

Answer 26

Cycle Service level: Measures the cycles without stockouts

Fill-rate: Measures the demand served from inventory

The CSL does not match the experience by the customer, since the total cycle is “failed” if only 1 customer experience stock-out.

Question 27

How do we measure lumpy demand?

Answer 27

Ratio = average demand / median demand

Higher ratio = more lumpy

Question 28

How to find re-order point for lumpy demand?

Answer 28

Simulations conducted in Excel will help