Poisson Probability Distribution Assignment
Scenario
A local bakery, known for its artisan bread, monitors operational data to improve its
business strategies. On weekdays, the bakery receives an average of 4 online bread orders
per hour. Some customers repeatedly attempt to order limited-edition bread until they
succeed. The bakery also conducts special promotions, randomly selecting a certain number
of customers to win free bread vouchers. Your task is to study this data using probability
distributions to help the bakery in its operations.
Question
Find the probability that the bakery will receive exactly 6 orders in a single hour on a
weekday, with clear indications of all notations, formulas, and calculation steps. Also,
discuss the key features of the Poisson distribution that make it suitable for this situation.
Answer
Step 1 — Define the random variable and parameter
Let X = number of online bread orders received in one weekday hour. Assume X follows a
Poisson distribution with mean (rate) λ = 4 orders/hour. Notation: X ~ Poisson(λ = 4).
Step 2 — Poisson probability formula
The probability that X = k is:
P(X = k) = (e^(-λ) * λ^k) / k!
We want P(X = 6), so set k = 6 and λ = 4:
P(X = 6) = (e^(-4) * 4^6) / 6!
Step 3 — Compute each piece (showing arithmetic)
4^6 = 4096
6! = 720
4096 / 720 = 256 / 45 ≈ 5.6889
e^(-4) ≈ 0.0183156
P(X = 6) ≈ 0.0183156 × 5.6889 ≈ 0.1042
Answer: P(X = 6) ≈ 0.1042 (about 10.42%)
Why the Poisson Distribution Fits This Scenario
1. Counts in a fixed interval: Poisson models the number of events (orders) in a fixed time
interval (one hour).
2. Events are approximately independent: Individual customer orders are reasonably
independent of each other.
3. Constant average rate: The bakery has a known average rate λ = 4 orders per hour on
weekdays.
4. No simultaneous multiple events in small intervals: Poisson assumes the chance of two or
more arrivals in a tiny time interval is negligible — reasonable for customer orders arriving
one at a time.
5. Mean = variance: For Poisson, E[X] = Var(X) = λ. That makes checking real data easy: if
observed variance is near the mean (≈4), Poisson is plausible.
Caveats: If orders tend to cluster (many customers ordering at once), or variance is much
larger than the mean, a different model may be more appropriate (e.g., negative binomial).