The task and original case
Build a support report that shows completed work and unresolved work separately. A fast average among resolved tickets can coexist with an old backlog. The portfolio artifact should let a reader see both facts without interpreting the unresolved ages as completed service times.
| Case element | Value or requirement |
|---|---|
| Inputs | Tickets resolved in 1 and 2 days; two still open at 10 and 20 days. |
| Build | A two-panel backlog report with an explicit censoring note. |
This is an original invented exercise. It is independent practice, with no employer assessment or open-vacancy claim. Download the inputs, attempt your own result, then open the separate answer key.
Data and worked inputs
Build a result another person can follow
Use one row per ticket. Keep status, resolved duration and age at the observation cutoff in distinct columns. The two completed tickets took one and two days. The other two are still open at ages ten and twenty days. A missing resolved duration for an open ticket is meaningful; it is not a numeric zero.
Filter to resolved tickets for the resolved-time mean. Separately count open tickets and summarize their ages. pandas documents explicit missing-value handling; use a status condition as well as a missingness check so that a malformed resolved record is not mistaken for an ordinary open ticket. Publish two panels: completed durations and the open-ticket register.
Sources[1]
Check the worked result
The completed-ticket mean is (1 + 2) ÷ 2 = 1.5 days. Two of four tickets remain open. Their observed ages are ten and twenty days, with a mean age of fifteen days at this cutoff. Those ages are not resolution durations because the closing events have not occurred.
Describe this as a snapshot. It does not establish an overall mean time to resolve all four tickets. The result gives a reviewer a concrete reason to investigate unresolved work even when the completed panel appears fast.
Make the failure visible
Averaging 1, 2, 10 and 20 produces 8.25 days, but mixes two different measurements. Filling missing resolution durations with zero yields 0.75 days and also answers the wrong question. Add a ticket marked resolved without a completion timestamp; send it to a data-quality exception list instead of including it in either panel silently.
Revise it and ask someone to reproduce it
Move the observation cutoff forward one day without closing the open tickets. Their ages should become eleven and twenty-one, while the resolved mean remains 1.5. Then close the oldest ticket and explain why it changes panel membership. Have a peer point to the cutoff and the unresolved cases before accepting your revision.
- The resolved mean uses only completed tickets.
- The open register retains status, age and observation cutoff.
- Missing or contradictory completion data receives an explicit exception.
Describe the work accurately
After completing your own version, you could adapt this project bullet: “Created a practice support report separating resolved durations from open-ticket ages and flagging contradictory status records.”
Use Projects or Independent practice as the heading. Keep the input files, result and revision together. Describe only your own completed work; the invented numbers are model outputs, not employer results or measured business impact.