Poisson distribution P(λ): Models rare events in fixed interval. P(X=k) = (e^(-λ) × λ^k) / k!. Mean = Variance = λ (unique property). Example: Defects per 1000 units, λ=3. P(X=2) = (e^(-3) × 3²)/2! ≈ 0.224. Assumptions: (1) Events independent; (2) Constant rate; (3) No clustering; (4) Rare events. Applications: Accidents/day, calls/hour, typos/page. Normal approximation: When λ > 5, use N(λ, √λ). Compared to binomial: Use Poisson when n large, p small, np = λ. Solving: Identify λ. Use formula or tables. Apply normal approximation if needed. Exam tip: Recognize Poisson context. Understand when Poisson vs binomial. Practice: Rare event problems.