91Ó°ÊÓ

The binomial distribution is a fundamental concept in probability and statistics. It describes the likelihood of a specific number of successes in a fixed number of trials, where each trial has only two possible outcomes: success or failure. In other words, it's like flipping a coin a certain number of times and counting how many times you get heads.

Here, a 'success' could mean many things. For instance, when considering the defect of bolts, a 'success' might mean a bolt is defective. The binomial distribution is characterized by two parameters:

• \( n \): The number of trials or experiments (in this case, 50 bolts in a box)
• \( p \): The probability of success on a single trial (here, a 3% chance for a bolt to be defective)

Whenever you have a large number of trials and a small probability of success, the binomial distribution can be approximated by the Poisson distribution, which simplifies the computation of probabilities.

The probability of defectives refers to calculating the likelihood of finding defective items in a batch. In our example with bolts, this indicates predicting how many out of 50 bolts could be defective. Assessing this gives manufacturers insights into quality control and aids in making data-driven decisions.

To calculate such probability, we look into scenarios like:

• Exactly zero defectives
• Exactly one defective
• Up to 5 defectives

These calculations help in understanding the quality production process. With a defect probability of 3%, and 50 bolts per box, manufacturers can predict defect rates and identify whether specific interventions are required to improve quality.

The Poisson approximation is a useful method when dealing with binomial distribution scenarios that involve a large number of trials (\( n \)) and a small probability of success (\( p \)). By converting the binomial problem to a Poisson problem, calculations become simpler and easier to manage.

Here's why Poisson is easier:

• Instead of calculating probabilities for each outcome as in binomial, you have a single parameter \( \lambda \) (lambda), which is the average rate of success across trials and is calculated as \( \lambda = n \times p \).
• The Poisson probability formula is \( P(X = x) = \frac{e^{-\lambda} \lambda^x}{x!} \), which requires fewer computations.

For example, when packaging bolts where each has a small chance of being defective, using Poisson helps to quickly estimate the probability of different numbers of defectives in a box of bolts. In the exercise, we use \( \lambda = 1.5 \) to approximate the probability of up to 5 defects, instead of dealing with repeated binomial calculations.