Understanding Lead Time

Question 1

What is the best way of answering the question of when will something be done?

Answer 1

The best way to answer is to use Lead times

Question 2

Is lead time a single value?

Answer 2

No lead time comes from a probability distribution.

It is a random number defined by a probability distribution function

Question 3

Why is lead time a probability distribution?

Answer 3

In a Kanban system, the duration from commitment to delivery is unpredictable and can vary.

Question 4

What can be visualised from a probability distribution?

Answer 4

Different distributions create curves that offer insight into the various work types and associated risks.

Question 5

Understanding the lead time curve helps with?

Answer 5

This helps us make smart risk-management and planning choices that lead to the best possible economic results, even with the uncertainty we face.

Question 6

How are lead time distribution functions used

Answer 6

They apply to practical challenges like predicting and controlling delay cost risks.

Question 7

Why is understanding lead time necessary?

Answer 7

Understanding lead time is essential because it aids in progressing to higher Kanban maturity.

Question 8

What is lead time?

Answer 8

1. Begins at the agreed commitment point. (Team starts work)
2. Ends when an item is ready for delivery. (Team Stops work)
3. The period of time that commitment must precede delivery

Question 9

What problem can occur with lead time commitment points

Answer 9

The commitment point is frequently unclear (ambiguous).

Question 10

What happens when the lead time commitment point is unclear (ambiguous)?

Answer 10

We measure lead time when the delivery team pulls the item into WIP.

Question 11

What are the different types of lead times

Answer 11

1. Customer lead time
2. System lead time

Question 12

What is customer Lead time?

Answer 12

The customer lead time is counted from the moment a customer makes a request to the point where that request is fulfilled and delivered to them.

Question 13

What is system Lead time?

Answer 13

System Lead time is the duration from when the delivery team chooses a customer request to work on to when it is delivered.

Question 14

What does synchronous commitment mean?

Answer 14

Customer lead time equals system lead time

Question 15

What happens when the Customer lead time is not equal to the system lead time

Answer 15

Customer lead time suffers from a longer, fatter tail

Question 16

What is used to visualise lead times?

Answer 16

histogram

Question 17

How is a lead time Histogram constructed

Answer 17

1. The x-axis shows discrete lead times in days for each completed customer request in ascending order.
2. The y-axis shows the frequency of lead times recorded on the x-axis, indicating the number of items processed through the kanban system within a specified time frame.

Question 18

How can lead time distributions be described?

Answer 18

1. Fat Tailed
2. Thin tailed

Question 19

What are the attributes of a fat-tailed distribution

Answer 19

A long, visible tail stretches out to the right along the X-axis.

Fat-tailed is undesirable and risky and makes planning difficult.

Question 20

What are the attributes of a thin-tailed distribution

Answer 20

There isn’t a long, visible tail running off to the right along the x-axis

Question 21

Why is knowing what type of distribution helpful?

Answer 21

Knowing which one you have in your Kanban system makes a vital difference in the following:

1. Planning.
2. Risk management.
3. Understanding the likelihood of achieving customer satisfaction.
4. Being viewed as a trustworthy service provider.

Question 22

What probability density functions are used with lead time distributions?

Answer 22

Weibull functions

Question 23

How do you know if you are viewing a Thin or Fat-tailed distribution?

Answer 23

see pic

Question 24

What are the key points on a distribution curve

Answer 24

1. Mode
2. Mean
3. Median
4. Tail (98th percentile)
5. 85th percentile ( 6 out of 7 Items).

Question 25

What is the Mode in a lead time distribution curve?

Answer 25

The mode is the top of the hill, the most commonly occurring lead time in the data set

Question 26

What is the Median in a lead time distribution curve?

Answer 26

1. The median is the 50th percentile.
2. half the lead times are before this value, and half are after that value.

Question 27

What is the mean in a lead time distribution curve?

Answer 27

1. The mean is the arithmetic average:

Sum up the value of all the data points and divide by the number of points.

Question 28

What affects the mean?

Answer 28

The mean typically diverges from both the mode and the median as the tail extends further to the right due to the influence of outliers.

A fat tail influences the mean far more than it influences the median and is unlikely to affect the mode at all.

Question 29

Why is it important to understand how the mean is affected by outliers

Answer 29

Planning is affected because simple forecasting equations, such as Little’s Law and regression to the mean, require using the mean.