Department Operations and Performance Improvement
Domain 3 tests whether you can run the department as a system: match capacity to demand, remove the constraint, control inventory, and prove improvement with data instead of opinion.
Capacity, flow, and the constraint
Define demand, capacity, and the bottleneck before changing anything. Little's Law: work in process = arrival rate × time in system, so shortening time means cutting WIP or the arrival rate. Turnaround time (TAT) should be decomposed into steps - order, collect, transport, analyze, verify, report - and measured per step; the longest step is usually not the analyzer.
Pick the right improvement method
- Model for Improvement: three questions (aim, measures, changes) plus PDSA cycles for rapid tests of change.
- DMAIC when the cause is unknown and the process is statistically stable.
- RCA (5 whys, fishbone) after a specific event.
- FMEA to predict failure in a new or redesigned process.
- Run and control charts: six consecutive points on one side of the median signals a shift; a single point is noise.
Inventory discipline
Par level = average use during lead time + safety stock for variability. Two-bin systems make reordering visual; assign expiry sweeps to a named owner and review slow movers.
Worked example
STAT chemistry TAT rises from 45 to 62 minutes. Step times: collection 8, transport 22, analysis 15, verification 17. Volume is up 20% and staffing is flat. The constraint is transport, not analysis, so buying a faster analyzer would not help. Options are more frequent courier runs or rerouted pneumatic tube service, followed by a run chart to confirm the shift.
Traps that cost marks
- Adding capacity at a non-constraint step.
- Reacting to one data point as though it were a trend.
- Reporting averages only - the tail is what patients and clinicians feel.
- Naming a person as the root cause of a process problem.
- Setting par levels from last month's order instead of use and lead time.