IOE 333 Lecture Notes - Lecture 21: Detection Theory

• School: U of M
• Department: Industrial And Operations Engineering
• Course: IOE 333
• Professor: Clive D'Souza

Document Summary

Model of how humans separate signal from noise. Need to gather information through senses and decide if there is enough evidence to constitute a signal. The world can be modeled as signal present or absent. World is assumed to contain noise that is random and normally distributed (noise is always present) Signal increases the mean of the noise distribution. How good operator is at separating signal from noise (bottom-up processing) Can be influenced by signal to noise ratio, redundancy of signal, operator alertness. Want to decrease overlap in the distributions. Bias/tendency for yes" and no" (top down processing) Impacts type of errors: false alarms, misses.