This assignment is completed as an in-class workshop.
Team Green has empirically estimated its velocity for their 3-week sprints at 47 story points. How long will it take to complete a 231 story point backlog. Assume that the backlog does not change and that velocity remains constant. What delivery date would you feel comfortable with assuming the project starts July 1.
Team Purple has an observed velocity for their 2-week sprints at 31 story points. How long will it take to complete a 329 story point backlog. Assume that the backlog does not change and that velocity increases 10% sprint-over-sprint for the next 7 sprints and then levels off.
Joanne Chung's Scrum team has been working diligently on their web portal project and has delivered solid features at the end of each 2-week sprint. She has observed velocities of 23, 31, 28, 41, 37, 39, and 44 points per sprint.
- Forecast velocity for the next sprint using a Simple Moving Average of the prior 3 sprints.
- Forecast velocity for the next sprint using a Weighted Moving Average of the prior 3 sprints with weights 4 for the most recent sprint, and 2 and 1 for the sprints prior to the last one, respectively.
What is the 95% confidence interval for the above velocity estimate? Recall that the confidence interval can be calculate as forecast +/- 1.96*SE, where SE = sigma / sqrt(n - 1) and sigma = 0.8*MAD.
Joanne Chung's Scrum team has been working diligently on their web portal project and has delivered solid features at the end of each 2-week sprint. She has observed velocities of 23, 31, 28, 41, 37, 39, and 44 points per sprint. Forecast velocity for the next sprint using a regression trend line. Which regression model fits best? What is the 95% confidence interval for the velocity estimate? Is this estimate better than the ones obtained in questions 3 and 4? How would you judge the estimates?
The scrum team has a backlog of 231 story points of work at the outset of the project. The following sprint goals were recorded: 31, 29, 38, 42. Create a backlog burndown chart. What is the meaning of the slope of the curve?
Team Yellow has been asked to build a release plan for the next 3 sprints. It does not have an empirically derived velocity yet, although the mostly same team had a mean velocity of 78 points/sprint on the last project. The team has been tasked to implement a learning management system similar to Blackboard or Schoology and has chosen a sprint length of 2 weeks. The following user stories have been identified and sized and are in order of importance:
- As an instructor, I want to post an assignment (9 points)
- As a student, I want to submit an assignment with a file attachment and a comment (6)
- As an instructor, I want to view an assignment submission (4)
- As an instructor, I want to record a grade with comments for an assignment submission (3)
- As a student, I want to view my grade for an assignment (4)
- Plus additional stories in order of importance with sizes of 2, 8, 4, 7, 12, 19, 11, 8, 2, 1, 4, 2, 9, 10, 7.
Team Yellow is using a Kanban system to map its workflow. It has observed the following lead times for the user stories produced by the Product Owner (link to data)
Plot the data and build a regression model that allows you to forecast lead time based on the size of a user story. Evaluate the fit of the model. Can you calculate a 95% confidence interval for the lead time estimates? What would the estimated lead time be for an 8 point story?