Case context: A team compares several microphone-based wake word detectors. Some exceed one false positive every 24 hours of operation. The remaining detectors meet that limit but differ in how often they miss a spoken wake word.

Question: How should the team screen and compare these detectors, and what property should determine the final choice?

Sample answer: The team should first reject any detector that exceeds one false positive every 24 hours because it fails the satisficing requirement. It should then compare the false negative rates of the remaining detectors and select the one with the lowest rate, since false negative rate is the optimizing metric.

Key points:

• Exclude detectors that violate the false-positive limit.
• Treat one false positive every 24 hours as the maximum acceptable frequency.
• Compare false negative rates only among qualifying detectors.
• Choose the qualifying detector with the lowest false negative rate.

Rubric: The response should apply the false-positive threshold as an eligibility constraint and then use minimum false negative rate to choose among eligible detectors.