Hold Baggage Alarm Rate That Nobody Owns
An airport's hold baggage screening system at a mid-sized terminal processes about 6,000 bags per day across four lines. Over three weeks, the rate of bags sent to the resolution area has risen from 8% to 19%. The resolution area has two operators on the day shift and one at night. Average resolution time has grown from 45 seconds to over 3 minutes, and 12 bags were loaded onto aircraft before resolution was completed, later discovered during reconciliation. Maintenance logs show one line has had an intermittent detector fault for 10 days, with a 'reset and continue' note. The screening supervisor says the increase is 'just a seasonal mix of dense bags' and that the operators are experienced.
1. Distinguish the plausible causes of the alarm-rate increase and describe how you would test each.
Read the worked answer
Separate causes into equipment, image/algorithm, object, and human factors. Equipment: the intermittent detector fault on one line is a prime suspect; test by comparing alarm rates per line, reviewing the fault history, and running a known test piece through each line. Algorithm or configuration: a recent software or threat-library update could raise sensitivity; check change logs and compare rates before and after any change. Object mix: seasonal dense bags can raise alarms, but a rise from 8% to 19% is too large to attribute to mix alone; test by sampling bag composition and comparing like-for-like. Human factors: operator decisions at the resolution point may be inconsistent, or operators may be clearing images they should escalate; test by auditing a sample of resolved images against a second reviewer. The reconciliation failures indicate a process control problem independent of the cause.
2. Twelve bags were loaded before resolution. What immediate actions and root-cause questions follow?
Read the worked answer
Immediate actions: identify the 12 bags and their flights, notify the relevant authority per the airport's security event procedure, and determine whether any bag is still accessible for resolution. Preserve records: images, operator identities, timestamps, and the loading sequence. Suspend the practice that allowed loading before resolution, and add a hard system or procedural hold so a bag cannot be released without a resolution record. Root-cause questions: was the hold overridden manually, and by whom; was the night shift understaffed relative to the queue; was the intermittent fault producing false clears; and did the reconciliation process detect the gap only after departure? The answers determine whether the fix is technical (system interlock), staffing (night coverage), or cultural (production pressure overriding security). All three should be addressed, with the interlock as the non-negotiable control.
3. The supervisor attributes the problem to seasonal bag mix. How do you challenge this respectfully and what data do you bring?
Read the worked answer
Acknowledge the hypothesis and test it rather than dismiss it. Bring per-line alarm rates for the same period in the previous year, the daily bag count and composition sample, the fault log for the affected line, and the distribution of resolution times by operator and shift. If the alarm rate rose on three lines without faults and on the faulted line more sharply, mix alone cannot explain it. If the rise coincides with a software change, that is stronger evidence. Present the reconciliation failures as the shared concern, since they are a fact rather than an interpretation. Frame the conversation around the common goal of reducing unnecessary alarms and eliminating premature release, and agree on a two-week data collection plan with defined owners so the conclusion is jointly owned.