Project Management Stack Exchange is a question and answer site for project managers. It only takes a minute to sign up.
Sign up to join this community
I'm working with a team (but are not part of the team) that is consistently going over their WIP limits.
How do I know if their WIP limit is realistic and if we need to change it?
Other than the WIP limit being violated, is the team experiencing other problems due to this issue? For example, has their lead or cycle time suffered during such events? Have you seen impact on upstream or downstream lanes - where work might seem to either pile up or dry up ( too much or too little)? Do you see an impact on team throughput?
In other words, is the flow of work impacted in any manner?
If cycle time has gone up or throughput gone down, that's an indication that WIP limits are realistic and they should not be exceeding them. If there is no impact and they are able to handle the extra work without impact, then perhaps the limits might be too low and need to be raised.
Another way to look at this might be to see how many cards each person is handling simultaneously. If they are more than 2-3 (which might also be too many!), the team is not taking the 'avoid multitasking' lesson of Kanban seriously - and it should impact Cycle time.
If you could convince them to - for a period of about 1-2 weeks - to strictly not exceed the current WIP limits - there again you might be able to show impact on cycle time - it should go down from current levels!
HTH
A good limit is basically one that is hit often, but not all the time.
"Consistently" could either mean the limit is too low, or could mean that it is good. How often is consistently?
If a limit is being hit once or twice a week for example, it sounds pretty good. I would probably have to ask why are they breaking it? Why is breaking the limit a better option that doing some of the alternatives, like storming on something already in work, fixing impediments or working on a different task?
I think to really get to the bottom of it I would have to hear what sorts of conversations are happening when that limit is hit.
I hate to jump in on a older thread, but this question (and an answer or two) seem to indicate a shallow understanding of what a WIP Limit is and is used for.
The WIP Limit policy is poorly named if you are doing anything more than a shallow kanban system implementation. We used the term WIP Limit (the actual name) originally because we were so overburdened that we needed language (specific words) about a policy that reduced overburdening. In that situation WIP Limit may be considered an appropriate name/label for the policy.
Deeper kanban implementations stop thinking of reducing overburdening (which is a significant and positive side effect) and prefer to think of economically optimizing flow. The term Optimal WIP starts becoming a more appropriate name for the policy.
Most people think that WIP policies are only intended to stop push, which they do, but they are also a signal to the producer to pull. If you are cronically under your WIP policy number, you should lower it.
As to the original question, the optimal WIP matches the arrival rate of work into a queue with the departure rate. You may need to step back and look at the overall system and not individual phases in your workflow to see this appropriately.
You may also be experiencing emotional resistance to a policy that would be beneficial if properly observed. I'd start with asking the team members why they aren't able to follow a policy that is intended to promote good behaviour.
I'd also add that WIP policies are not laws. They are guidance to what a team thinks is good behaviour. Exceeding them is fine as long as the decision to make an exception is informed and actually an exception and not the norm.
Check out my blog post for a more complete introduction to this idea.
If your WIP limit is within plus or minus two or three sigma of your actuals, it is realistic. You need to change it if it falls outside of that. If you want more assurance that they will hit it, then raise the WIP limit to a more pessimistic value, but your planned costs will have to climb accordingly.
