Lesson 8: Modelling with exponential random variables
Choosing a warranty period
SkillsModelling waiting timesExponential quantiles
Worked example
The lifetime of a cordless drill is modelled as exponential with a mean of 6 years. The manufacturer wants to offer a warranty, in whole months, under which at most 5 % of drills fail.
a) Determine the longest warranty the manufacturer can offer.
b) Evaluate the suitability of the exponential model for this decision.
Step 1: Define the variable
Let be the lifetime in years, .
Step 2: Write the condition
For a warranty of years, the proportion failing within it is . We need .
Step 3: Solve
which is months.
Multiplying by reverses the inequality.
Step 4: Round in context
A 4-month warranty would let more than 5 % fail, so round down: the longest warranty is 3 months.
Step 5: Evaluate the model
An exponential model has no memory, so it assumes a drill is as likely to fail in its first month as in its sixth year. Real tools have some early failures, then a long reliable period, then wear out. Under this model 5 % of drills fail within about 3.7 months, which seems pessimistic for a product with a 6-year mean life; with real failure data the manufacturer could probably offer a longer warranty.
Answer
a) 3 months (the exact bound is about 3.69 months). b) The model assumes a constant failure rate with no wear-out, which is unrealistic for drills; real data would give a better answer.
Try a similar one
A fresh question of the same kind, checked as you answer, with hints and a full solution.
Method
Define with units and .
Turn the requirement into an inequality in .
Solve with logarithms, keeping track of the inequality's direction.
Round in the direction the context requires.
Evaluate the constant-rate assumption.
Watch out for
Rounding 3.69 months up to 4 months, which breaks the 5 % limit.
Solving , which finds when 95 % have failed.