As I mentioned in a previous post, I've been experimenting with using the I saw David Anderson mention Kanban System to test how this would work in our organization.  As mentioned earlier the paper A Kanban System for Sustaining Engineering explains This method.  See my previous post for an explanation of what we are attempting. 

I wanted to share a couple of issues I've encountered and how I deal with them. First, we have set up the system so that we are not accepting any more changes until the current batch are complete.  The biggest plus so far has been the huge visibility to management.  We can see what tasks are still to do, and we were able to determine that a certain type of task with only one developer had over half the issues.  We were able to get resources to back this other resource up, which is helpful

The other issue that affects this system are what I call "emergency changes" that ensue from issues around the User Interface.  For instance if you are using a technology and having problems with that technology how is that captured?  Right now we're treating that not as a change, but a problem.   I'd like to hear if anyone else is doing anything else with this.

Happy PMing/Coding!

Share this post: email it! | bookmark it! | digg it! | reddit! | kick it! | live it!
Sponsor