Example 1
Question: A factory has three machines X, Y, and Z producing bolts. Machine X produces 2% defective bolts, machine Y produces 4% defective bolts, and machine Z produces 6% defective bolts. Machine X is in operation 40% of the time, machine Y is in operation 35% of the time, and machine Z is in operation 25% of the time. A bolt selected at random is found to be defective. Find the probability that it was produced by machine Y.
Solution: Let \( E_1, E_2, E_3 \) be the events that a bolt is produced by machines X, Y, Z respectively, and let \( D \) be the event that the bolt is defective. The prior probabilities are: \[P(E_1) = 0.40, \quad P(E_2) = 0.35, \quad P(E_3) = 0.25\] The conditional probabilities of producing a defective bolt are: \[P(D \mid E_1) = 0.02, \quad P(D \mid E_2) = 0.04, \quad P(D \mid E_3) = 0.06\] Using the Total Probability Theorem, the probability that a randomly selected bolt is defective: \[P(D) = P(E_1)\cdot P(D \mid E_1) + P(E_2)\cdot P(D \mid E_2) + P(E_3)\cdot P(D \mid E_3)\] \[P(D) = (0.40)(0.02) + (0.35)(0.04) + (0.25)(0.06)\] \[P(D) = 0.008 + 0.014 + 0.015 = 0.037\] By Bayes' Theorem, the probability that the defective bolt was produced by machine Y is: \[P(E_2 \mid D) = \frac{P(E_2) \cdot P(D \mid E_2)}{P(D)}\] \[P(E_2 \mid D) = \frac{(0.35)(0.04)}{0.037} = \frac{0.014}{0.037}\] \[P(E_2 \mid D) = \frac{14}{37}\]
Answer: \(\dfrac{14}{37}\)
Example 2
Question: In a call centre, three employees P, Q, and R handle customer calls. Employee P handles 45% of the calls, employee Q handles 30%, and employee R handles the remaining 25%. The probability that a call handled by P results in a complaint is 3%, by Q is 5%, and by R is 8%. A complaint call is randomly selected. What is the probability that it was handled by employee R?
Solution: Let \( E_1, E_2, E_3 \) be the events that a call is handled by P, Q, R respectively, and let \( C \) be the event that the call results in a complaint. The prior probabilities are: \[P(E_1) = 0.45, \quad P(E_2) = 0.30, \quad P(E_3) = 0.25\] The conditional probabilities of a complaint are: \[P(C \mid E_1) = 0.03, \quad P(C \mid E_2) = 0.05, \quad P(C \mid E_3) = 0.08\] Using the Total Probability Theorem: \[P(C) = P(E_1)\cdot P(C \mid E_1) + P(E_2)\cdot P(C \mid E_2) + P(E_3)\cdot P(C \mid E_3)\] \[P(C) = (0.45)(0.03) + (0.30)(0.05) + (0.25)(0.08)\] \[P(C) = 0.0135 + 0.015 + 0.020 = 0.0485\] By Bayes' Theorem, the probability that the complaint call was handled by R is: \[P(E_3 \mid C) = \frac{P(E_3) \cdot P(C \mid E_3)}{P(C)}\] \[P(E_3 \mid C) = \frac{(0.25)(0.08)}{0.0485} = \frac{0.020}{0.0485}\] \[P(E_3 \mid C) = \frac{20}{48.5} = \frac{40}{97}\]
Answer: \(\dfrac{40}{97}\)