1
Name: Muhammad Usman
Roll no: BS-AI 240812
Semester: 4th
Assignment: Probability and Statistics
INTRODUCTION TO PROBABILITY DISTRIBUTIONS
In today's data-driven world, multinational companies (MNCs) generate enormous volumes of
data every second from customer transactions and server logs to product demand and supply
chain events. Probability distributions are the mathematical backbone that helps businesses
make sense of this uncertainty. They describe how likely different outcomes are in a process,
enabling companies to forecast, plan, and make smarter decisions. Whether Amazon is
predicting the number of orders in an hour, Google is modeling server response times, or a bank
is assessing financial risk, probability distributions are working behind the scenes. This
assignment explores four fundamental probability distributions: Uniform, Poisson, Normal, and
Exponential, with their definitions, mathematical formulas, and real-world applications in globally
recognized companies.
Uniform Poisson Normal Exponential
Equal chance for all Count of rare discrete Bell-curve of Time between
outcomes events continuous data random events
1.UNIFORM PROBABILITY DISTRIBUTION
Every outcome has an equal likelihood of occurring.
Definition:
A Uniform Distribution is a type of probability distribution where all outcomes within a defined
range are equally likely. It comes in two forms: Discrete Uniform (finite equally-likely outcomes,
e.g., rolling a die) and Continuous Uniform (any value in an interval [a, b] is equally probable).
Mathematical Formula:
Continuous Uniform f(x) = 1 / (b - a) for a <= x <= b
Mean (µ) µ = (a + b) / 2
Variance (σ²) σ² = (b - a)² / 12
Key Characteristics:
• All outcomes in the range [a, b] are equally likely; no outcome is favored.
• The distribution is completely flat (rectangular shape when graphed).
• Used when there is no prior knowledge about which outcome is more probable.
• Both discrete (whole numbers) and continuous (any real value) forms exist.
2
Real-World Applications in MNCs:
Company Industry Practical Application
Amazon E-Commerce / Tech Randomly assigning delivery
time windows to customers
when all slots have equal
availability. Also used in A/B
testing to randomly assign users
to control or test groups with
equal probability.
Google Technology Random number generation in
cryptographic security protocols.
Uniform distribution ensures no
bias in key generation
algorithms.
Visa / Mastercard Financial Services Fraud detection simulations,
generating uniformly random
synthetic transaction data to
train and test fraud detection ML
models.
McDonald’s Fast Food / Retail Randomly selecting audit
locations across all restaurant
branches ensures that each
branch has an equal probability
of being chosen for quality
inspection.
Netflix Streaming / Tech Random sampling of user
sessions during system load
testing to simulate equal traffic
distribution across servers.
2.POISSON PROBABILITY DISTRIBUTION (DISCRETE)
Models the number of times a rare event occurs in a fixed time or space.
Definition:
The Poisson Distribution is a discrete probability distribution that describes the probability of a
given number of events occurring in a fixed interval of time or space, assuming these events
occur at a known average rate (λ, lambda) and are independent of each other. It is ideal for
counting rare, random events.
Mathematical Formula:
Probability formula P(X = k) = (e^(-λ) × λ^k) / k!
Where: λ = average rate of events | k = actual number of events | e ≈
2.71828
3
Mean and Variance Mean = λ and Variance = λ (both are equal)
Key Characteristics:
• Events must occur randomly, independently, and at a constant average rate.
• The variable k must be a non-negative integer (0, 1, 2, 3, ...).
• Uniquely, the mean and variance are both equal to λ.
• Works best for rare events, small probability over a large number of trials.
Real-World Applications in MNCs:
Company Industry Practical Application
Amazon Web Services Cloud Computing Predicting the number of server
requests per second to
auto-scale infrastructure. If the
average is 500 requests/sec
(λ=500), Poisson helps estimate
surge probabilities.
Uber / Lyft Ride-Sharing Estimating the number of ride
requests in a given zone per
10-minute window. Helps
pre-position drivers in
high-demand areas.
Pfizer / AstraZeneca Pharmaceuticals Modeling the number of adverse
drug reactions reported per
100,000 patients in clinical trials
for safety evaluation.
Deutsche Telekom Telecommunications Predicting the number of call
arrivals at a call center per
minute to optimize staffing levels
and reduce customer wait time.
DHL / FedEx Logistics Modeling the number of
package delivery failures per
route per day, enabling proactive
resource allocation and quality
control.
3.NORMAL PROBABILITY DISTRIBUTION
(CONTINUOUS)
The famous bell curve is the most widely used distribution in statistics
Definition:
4
The Normal Distribution (also called the Gaussian Distribution) is a continuous probability
distribution that is symmetric around its mean (µ), forming the classic bell-shaped curve. Most
values cluster around the central mean, and probabilities for values farther from the mean taper
off equally on both sides. It is the most important distribution in statistics, thanks to the Central
Limit Theorem.
Mathematical Formula:
Probability Density f(x) = [1 / (σ√(2π))] × e^[-(x-µ)² / (2σ²)]
Parameters µ = mean (center of bell curve) σ = standard deviation
(spread)
68-95-99.7 Rule 68% within 1σ | 95% within 2σ | 99.7% within 3σ
68% 95% 99.7%
Within ±1σ of the mean Within ±2σ of the mean Within ±3σ of the mean
Key Characteristics:
• Perfectly symmetric mean, median, and mode are all equal.
• Defined by just two parameters: µ (mean) and σ (standard deviation).
• The total area under the bell curve always equals 1 (100% probability).
• Foundation of many statistical tests: Z-test, T-test, ANOVA, and regression.
Real-World Applications in MNCs:
Company Industry Practical Application
Toyota / BMW Automotive Mfg. Quality control in manufacturing,
measuring the diameter of
engine parts. Parts within ±2σ
are accepted; outside are
rejected, ensuring precision at
scale.
JPMorgan Chase Banking / Finance Risk management and
Value-at-Risk (VaR) modeling.
Stock return distributions are
approximated as normal to
estimate potential losses over
time.
Google / Meta Technology / HR Modeling employee
performance scores and salaries
for compensation benchmarking
and identifying top/bottom
performers.
5
Coca-Cola FMCG / Beverage Monitoring beverage fill volumes
on production lines. The fill
quantity follows a normal
distribution; deviations trigger
machine recalibration.
Pfizer Pharmaceuticals Drug dosage response analysis:
patient responses to a medicine
are normally distributed,
allowing accurate
dose-effectiveness modeling.
4.EXPONENTIAL PROBABILITY DISTRIBUTION
Models the time between consecutive random events in a Poisson process
Definition:
The Exponential Distribution is a continuous probability distribution that describes the time
between events in a Poisson process, i.e., when events happen continuously and independently
at a constant average rate (λ). It is commonly used to model waiting times, service times, and
equipment lifetimes. A key property is memorylessness: the probability of an event in the next
moment does not depend on how long you have already waited.
Mathematical Formula:
Probability Density f(x) = λ × e^(-λx) for x >= 0
Cumulative (CDF) F(x) = 1 - e^(-λx)
Mean & Variance Mean = 1/λ | Variance = 1/λ²
Key Characteristics:
• Describes the waiting time between events (e.g., time until next customer arrives).
• Memoryless property: P(X > s+t | X > s) = P(X > t) the past has no influence.
• Closely related to the Poisson distribution if arrivals follow Poisson(λ), inter-arrival times follow
Exponential(λ).
• Always right-skewed, most events happen quickly, and few take very long.
Real-World Applications in MNCs:
Company Industry Practical Application
Amazon / Flipkart E-Commerce Modeling the time between
customer orders on the platform
to optimize warehouse staffing,
picker allocation, and dispatch
scheduling in real time.
Airbus / Boeing Aerospace / Mfg Predicting time-to-failure for
6
aircraft components and jet
engines using exponential
distribution models to determine
the lifespan of components for
scheduled maintenance
planning.
AT&T/ Vodafone Telecommunications Modeling call inter-arrival times
to design efficient queuing
systems and minimize wait
times in customer support and
network routing.
Netflix / YouTube Streaming Tech Analyzing session lengths and
time between user interactions
(clicks, pauses) to optimize
recommendation algorithms and
content delivery network (CDN)
performance.
Siemens Industrial / IoT Reliability engineering models
the time between equipment
failures in manufacturing plants
to plan preventive maintenance
and reduce unplanned
downtime.