Five Problems, Forty Engineering Hours
A shared IT service supports 12,000 staff across three hospitals. The problem manager has 40 hours of senior engineer time this month and five candidate problems. Problem A: an interface drops about 0.8% of radiology orders; clinical staff re-enter them manually, average 4 minutes per order, no patient harm yet. Problem B: the payroll export fails once per quarter, requiring 6 hours of manual reconciliation by finance; the next run is in 10 days. Problem C: a legacy scheduling module has a known error that causes duplicate bookings at one small clinic, 2–3 times a week. Problem D: a database index is degrading, increasing report runtime from 2 to 9 seconds for 300 users; no outage. Problem E: a new CIO has asked for 'root cause' of a single executive's VPN failure last month. Impact data is incomplete for A and D. How should the problem manager prioritize?
1. Rank or sequence the five problems using ITIL problem prioritization concepts. State your assumptions and how you would handle the data gaps.
Read the worked answer
Sequence the work as follows, and state assumptions. B first: the quarterly payroll deadline is 10 days away, failure has a hard financial control impact, and the cause is likely to be reproducible. A second: 0.8% of radiology orders at 12,000 staff is a meaningful clinical workflow burden, with a rising trend and possible patient-safety risk; gather quick data before committing full effort. D third: 300 users losing 7 seconds per report is real productivity loss but lower risk and can be batched. C fourth: localized duplicate bookings, low volume, good candidate for a known-error workaround. E last: one executive's VPN failure is an incident unless it recurs; check history before opening a problem. Use impact, urgency, risk, and trend, not seniority. Re-rank if A shows patient harm or a sharp increase.
2. A clinician emails the CEO about Problem A, and the CEO asks why Problem B is being worked first. Draft a response explaining prioritization without dismissing clinical risk.
Read the worked answer
Acknowledge the clinical concern first, then explain the criteria: a near-term financial control deadline, frequency, risk, and the cost of delay. Do not say B is more important than patient care. Explain that A is being scoped now with a short data request and an interim workaround, and set a review date. Offer to show the comparison and ask the clinician for evidence of harm or workaround failure. If A is causing delayed care, re-prioritize immediately. The response should be transparent, specific, and owned by the problem manager, not delegated to the service desk. This protects trust and keeps prioritization evidence-based rather than escalation-based.
3. What should the problem manager do about Problem E?
Read the worked answer
For E, first check whether the VPN failure is a one-off or part of a pattern. A single incident should not become a problem just because an executive is affected. If there are related failures across the executive team, group them into a problem with a clear statement and impact. If not, close with an explanation and point to the standard incident process. This is not about ignoring the CIO; it is about applying the same definition and evidence standard. Document the decision so the criteria are visible. The problem manager should also watch whether repeated VIP escalations are distorting the backlog, and raise that as a governance issue.