Applications: Reliability and Survival Functions
You'll learn how to use cumulative distribution functions (CDFs) to model and analyze lifetimes, reliability, and survival in practical scenarios.
The Core Idea
When analyzing how long something lasts—whether it's a machine component, a light bulb, or even time until an event in medical studies—we often care about the probability that it survives beyond a certain time rather than fails by that time.
The survival function (also called the reliability function) is defined as:
where F(x) is the CDF of the lifetime random variable X.
What Does It Mean?
- F(x) = P(X ≤ x) = probability of failure by time x
- S(x) = P(X > x) = probability of survival beyond time x
Think of a light bulb: if X represents its lifetime in hours, then F(1000) tells you the probability it burns out within 1000 hours, while S(1000) tells you the probability it's still working after 1000 hours.
Why It Matters
In reliability engineering and survival analysis, we naturally think about "surviving" rather than "failing." The survival function gives us this perspective directly:
- S(0) = 1: Everything starts working (100% survival at time 0)
- S(∞) = 0: Eventually everything fails
- S is decreasing: Survival probability decreases over time
This function is central to warranty analysis, maintenance scheduling, actuarial science, and clinical trials.
Key Takeaway: The survival function S(x) = 1 - F(x) transforms the CDF into a probability-of-survival perspective, making it the natural tool for reliability and lifetime analysis.