WIP and lead time (also called delivery time)

Transferred from: Http://www.jianshu.com/p/bbc652391e45

In the product and the pre-set time is basically linear relationship, reducing the number of WIP can reduce the pre-time.

The law of Little, as a very simple principle, lays a theoretical foundation for the Kanban method, and the seemingly simple formula has a complex side behind it. first, the law of Zermatt

The formula for the law of Delphi is this:

Average throughput = WIP quantity/Average pre-set time

For example, suppose you are queuing up for fast food, there are 19 people in front of you in line, you are the 20th one, and you know that the Cashier window can handle a person's order demand every minute, solving your waiting time.

If you have decided to queue and stand at the end of the line, then the number of products is 20 (each), the average throughput rate is 1 (person/minute).
From the time you stand to the end of the line, until you finish your meal, this time is your "pre-set time".
Even if we have not learned the law of Delphi, we can easily calculate it:

    1 = 20/x
    x = 20 (minutes)

Because in a period of time, to maintain a full workload, we can do a lot of work every day basically is certain, so the throughput rate basically does not change too much.
If at this time we want to shorten the average lead time, which is the waiting time, the law of Delphi tells us that this goal can be achieved by reducing the number of WIP.
In this case, the number of people queuing is reduced.

It is also well understood that the queue of 10 people and the queue of 20 people, the former need to wait for a shorter time. [1]
Ii. The significance of limiting the product

As mentioned above, in the number of products and lead time is proportional to shorten the lead time is the most effective means to reduce the number of products.

The increase in lead times leads to longer delivery cycles, which is largely beyond doubt.

The increase in lead time leads to a decline in the predictability of delivery, as the saying goes, "nights", long periods of stay at a certain stage brings some additional risk.

If our delivery cycle is longer than the demand change cycle, there will be more urgent tasks, so a longer lead time will result in more urgent tasks.

If we do not manage the insertion of urgent tasks, we will increase the number of WIP.

If the delivery team's predictability is low, it will affect the trust relationship between the IT development organization and the business unit, and when the business unit cannot predict when a requirement will be delivered to the research and development department, then the only viable means is to put all the work to the research and development department at once. Directly increase the number of products in the research and development department.

At the same time, the increase in the number of products will bring another result is the failure to find the late, this point in the past thirty or forty years in the software engineering methodology has been verified.

The discovery of a failure requires resources and time to repair, resulting in the increase in the number of products and lead time.