CHAPTER 1: INTRODUCTION TO QUANTITATIVE ANALYSIS o Formula: BEP = Fixed cost ÷ (Selling price

1. Introduction – Variable cost)

• Mathematical tools have been used for thousands o Example (Pritchett’s Precious Time
of years. Pieces):
• Quantitative analysis applies to many real-world ▪ s = 10, f = 1,000, v = 5 → BEP = 200
problems. units
• It is not enough to know math—you must 6. Advantages of Mathematical Modeling
understand when and how to use it, its • Accurately represents reality
limitations, and assumptions. • Helps formulate problems
• Example: Taco Bell, NBC, and Continental • Gives insight and saves time/money
Airlines saved millions using quantitative models. • Only way to solve large/complex problems
2. What is Quantitative Analysis? • Communicates problems/solutions clearly
• Definition: A scientific approach to managerial 7. Role of Computers and Spreadsheets
decision making where raw data → meaningful • QM for Windows – decision support tool for
information. POM/QM.
• Factors: • Excel QM – works directly in spreadsheets;
o Quantitative factors (measurable): includes break-even, goal seek, etc.
investment alternatives, interest rates, 8. Possible Problems in Quantitative Analysis
inventory levels, demand, labor cost. • Defining the problem: conflicting views, wrong
o Qualitative factors (harder to measure): assumptions.
weather, legislation, technological • Developing a model: trade-off between
breakthroughs. complexity & understanding.
3. The Quantitative Analysis Approach (7 Steps) • Data collection: inaccurate or unavailable data.
1. Defining the Problem – Most important step; • Solution: math too complex, single solution
focus on true causes, not just symptoms. limiting.
2. Developing a Model – Build mathematical • Implementation: resistance, lack of
representation of the problem. commitment, “quick and dirty” mentality.
3. Acquiring Input Data – Must be accurate; GIGO 9. Implementation (Not Just the Final Step)
rule. • Requires management support + user
4. Developing a Solution – Use equations, trial and involvement.
error, enumeration, or algorithms.
• Analysts must be committed, not detached.
5. Testing the Solution – Validate with new data;
• Must monitor after implementation (conditions
results must be logical and realistic.
change).
6. Analyzing the Results – Study implications; use
Chapter 1: Introduction to Quantitative Analysis —
sensitivity analysis.
Expanded Reviewer
7. Implementing the Results – Apply in real life;
1. Introduction
difficult due to resistance to change.
• Mathematical tools have been used for thousands
4. Modeling
of years.
• Models are: realistic, solvable, understandable.
• Quantitative analysis can be applied to a wide
• Contain:
variety of problems.
o Variables (controllable = decision
• It’s not enough to just know the mathematics of a
variables, uncontrollable = environment).
technique. One must understand the specific
o Parameters (known constants like
applicability of the technique, its limitations, and
costs).
its assumptions.
• Types:
Examples in real life:
o Deterministic models – values known
• Taco Bell used forecasting and scheduling
with certainty.
models to save $150 million.
o Probabilistic models – involve
risk/uncertainty. • NBC used quantitative analysis to increase
5. Quantitative Model Example: Profit revenue by $200 million.
• Formula: Profit = Revenue – Expenses • Continental Airlines saved $40 million by using
models to recover from weather delays.
• Expanded: Profit = sX – f – vX
Recitation use: “Even fast-food and airlines rely on these
o s = selling price/unit
tools, not just math classes. For example, if McDonald’s
o f = fixed cost
predicts wrong customer traffic, they waste food and
o v = variable cost/unit
money. Forecasting models help them plan correctly.”
o X = units sold
• Break-even point (BEP):
2. What is Quantitative Analysis? o A pizza chain testing different delivery
• Definition: A scientific approach to managerial routes (trial and error) until it finds the
decision making in which raw data are processed fastest one.
and manipulated to produce meaningful 5. TESTING THE SOLUTION
information. o Verify accuracy; collect new data; results
Quantitative vs. Qualitative Factors must be logical and realistic.
• Quantitative factors (measurable): investment Example:
alternatives, interest rates, inventory levels, o A mall predicts foot traffic; testing means
demand, labor cost. checking if actual visitor counts match
• Qualitative factors (hard to measure): the the model.
weather, state/federal legislation, technological 6. A NALYZING THE RESULTS
breakthroughs. o Study implications before
Example: implementation.
• A retail store deciding how many shirts to stock: o Use sensitivity analysis = see how
o Quantitative = sales data, cost per shirt. results change with changes in input.
o Qualitative = fashion trends, celebrity Example:
influence, weather (e.g., jackets sell more o A bookstore predicts profit if it sells 100
in rainy months). books/week. Sensitivity analysis asks:
Recitation use: “Numbers like sales per week are “What if sales drop to 80 books?” or
quantitative, but a sudden trend on TikTok is qualitative. “What if supplier raises costs by
Both matter in making decisions.” ₱20/book?”
3. The Quantitative Analysis Approach (7 Steps) 7. IMPLEMENTING THE RESULTS
1. DEFINING THE PROBLEM o Put solution into practice. Often difficult
o Develop a clear and concise statement. due to resistance to change.
o Go beyond symptoms, find true causes. o Must be monitored over time because
o Select the right problems (focus). conditions change.
o Specific, measurable objectives are Example:
needed. o A bank installs an online queuing system.
Example: Even if it works, staff may resist, or
o A coffee shop with “low sales” might customers may need training.
think the problem is high prices, but the Recitation use: “Implementation is like launching a new
real cause could be poor location or lack system at Jollibee—it might work on paper, but employees
of marketing. need to accept it, or else the plan fails.”
2. DEVELOPING A MODEL 4. Mathematical Models: Profit & Break-even
o Models are realistic, solvable, • Profit = Revenue – Expenses
understandable. • Expanded: Profit = (Selling price × units sold) –
o Contain variables [Fixed cost + (Variable cost × units sold)]
(controllable/uncontrollable) and • Simplified: Profit = sX – f – vX
parameters (known quantities). Where:
Example: • s = selling price per unit
o In Grab or Uber, the controllable variable • f = fixed cost
= number of cars dispatched. • v = variable cost per unit
o Parameter = fixed cost of maintaining the • X = number of units sold
app. Break-even Point (BEP)
3. A CQUIRING INPUT DATA • Formula:
o Must be accurate → GIGO rule (“Garbage BEP = Fixed cost ÷ (Selling price – Variable cost)
In, Garbage Out”). • Example (Pritchett’s Precious Time Pieces):
Example: o s = 10, f = 1,000, v = 5 → BEP = 200 units
o If a grocery store miscounts its inventory, Real-life example:
the reorder model will suggest wrong • A milk tea shop spends ₱50,000/month (fixed
quantities → leading to shortages or cost). Each drink costs ₱30 to make and sells for
spoilage. ₱70.
4. DEVELOPING A SOLUTION o BEP = ₱50,000 ÷ (70 – 30) = 1,250 cups.
o Techniques: equations, trial and error, • If they sell fewer than 1,250 cups → loss. More
complete enumeration, algorithm. than 1,250 → profit.
Example: Recitation use: “Break-even tells us the safety line. If
Starbucks doesn’t hit break-even, it’s losing money even if
it’s selling drinks.”
5. Advantages of Mathematical Modeling CHAPTER 1: INTRODUCTION TO QUANTITATIVE ANALYSIS
• Accurately represents reality Memory Guide
• Helps formulate problems 1. Quantitative Analysis Basics
• Gives insight and information Definition: Scientific approach to decision making → Raw
• Saves time and money Data → Info
• Only way to solve complex problems quickly Code: SRI = Scientific → Raw → Info
• Communicates problems/solutions • Quantitative factors: M-I-I-D-L → Money
Example: (investment), Interest, Inventory, Demand, Labor
• Airline pricing models test different ticket prices • Qualitative factors: W-L-T → Weather,
before applying them in real flights → saves Legislation, Technology
millions without trial-and-error in real life. 2. Quantitative Analysis Approach (7 Steps)
6. Categories of Models by Risk Mnemonic: D-M-A-D-T-A-I
• Deterministic models: no risk, values are certain. Define problem
• Probabilistic models: involve risk, chance, Model (develop)
uncertainty. Acquire data
Example: Develop solution
Test solution
• Deterministic: A jeepney’s fuel cost per liter
Analyze results
(known value).
Implement results
• Probabilistic: Number of passengers per trip
Shortcut keyword: “Don’t Make A Dumb Theory About
(depends on chance).
Issues.”
7. Computers and Spreadsheet Models
3. Defining the Problem
• QM for Windows: decision support tool for
CLEAR
POM/QM.
• Concise statement
• Excel QM: works directly within Excel (break-
• Look past symptoms → real causes
even, goal seek, etc.).
• Essential problems only
Example:
• Aim: measurable objectives
• A bakery can use Excel QM to calculate profit at
different sales volumes without manually • Right problem selection
recomputing each scenario. 4. Developing a Model
8. Possible Problems in Quantitative Analysis • Must be: RSU → Realistic, Solvable,
1. Defining the problem: wrong assumptions, Understandable
conflicting views. • Has: VP → Variables & Parameters
2. Developing a model: too complex or • Variables: Controllable (decisions) vs.
oversimplified. Uncontrollable (environment)
3. Acquiring data: inaccurate or incomplete data. 5. Acquiring Input Data
4. Developing solution: math too difficult, or only GIGO Rule = Garbage In → Garbage Out
one rigid answer. • Sources: R-D-I-M-S = Reports, Documents,
5. Implementation: resistance, lack of Interviews, Measurement, Sampling
commitment, quick-fix mentality. 6. Developing a Solution
Example: STEA
• A clothing brand might build a perfect sales • Solve equations
forecast model, but if management ignores it and • Trial and error
sticks to intuition, implementation fails. • Enumeration (complete)
Recitation use: “Even the best formula is useless if people • Algorithm
won’t use it. That’s why implementation is key.” 7. Testing the Solution
9. Implementation – Not Just the Final Step “T-N-L-R”
• Requires management support + user • Test input/model
involvement. • New data collection
• Analysts should care about the problem and • Logical results
solution. • Real situation fit
• Continuous monitoring needed because 8. Analyzing Results
situations change. “SIS” = Study impact, Implement change, Sensitivity
Example: analysis
• A university installs an online enrollment system. • Sensitivity Analysis: “What if?” test → How much
It works at first, but when enrollment numbers results change if data/model changes
double, the system crashes → needs adjustment. 9. Implementing Results
HARD (why it’s difficult)
 • Human resistance • RIISCC (Model advantages)
• Analyst/management lack of commitment • DP (Deterministic vs. Probabilistic)
• Reluctance to lose decision power • Problems Dog My Decisions Till All
• Desire for quick & dirty methods Implemented (Problems list)
10. Profit & Break-even Model
Formula: Chapter 3: Decision Analysis
• Profit = Revenue – Expenses
Introduction
• Profit = sX – f – vX
• BEP = f ÷ (s – v) • Decision theory → Analytic and systematic
Remember: approach to decision making.
• s = selling price • Good decision → Based on logic, considers all
• f = fixed cost available data and alternatives, uses quantitative
approach.
• v = variable cost
• X = units sold Six Steps in Decision Making
BEP Mnemonic: “Fixed divided by Margin.” 1. Clearly define the problem.
(Margin = Selling Price – Variable Cost) 2. List possible alternatives.
11. Advantages of Models 3. Identify possible outcomes (states of nature).
RIISCC 4. List the payoff (profit) for each alternative-
• Represent reality outcome combination.
• Insight 5. Select a mathematical decision theory model.
• Information 6. Apply the model and make a decision.
• Saves time/money Example: Thompson Lumber Company deciding on large
• Complex problems solved plant, small plant, or no plant depending on market.
• Communicate solutions
12. Types of Models Types of Decision-Making Environments
DP = Deterministic (certainty), Probabilistic (risk) 1. Certainty → Consequences are known.
13. Possible Problems 2. Uncertainty → Outcomes known, but probabilities
D-M-D-T-A-I (same as steps, but problems in each) unknown.
• Define problem: unclear, conflicting, wrong 3. Risk → Outcomes known, and probabilities
assumptions known.
• Model: too complex/simple Decision Making Under Uncertainty
• Data: missing or invalid
• Maximax (optimistic) → Choose alternative with
• Solution: math too hard, only one answer maximum possible payoff.
• Analyze/Test: incomplete view • Maximin (pessimistic) → Choose alternative with
• Implement: resistance, lack of commitment best “worst case.”
Shortcut phrase: “Problems Dog My Decisions Till All • Criterion of Realism (Hurwicz) → Weighted
Implemented.” compromise between optimism (α=1) and
14. Tools pessimism (α=0).
• QM for Windows → Decision support system • Equally Likely (Laplace) → Average payoff,
• Excel QM → Inside Excel, has Break-even & Goal choose highest.
Seek • Minimax Regret → Minimize maximum regret
(opportunity loss).
Quick Recap in Mnemonics Only
• SRI = Scientific → Raw → Info (Definition) Decision Making Under Risk
• M-I-I-D-L (Quantitative factors), W-L-T • Uses probabilities for outcomes.
(Qualitative factors) • Expected Monetary Value (EMV) = Σ (Payoff ×
• DMA-DTAI (7 steps) → “Don’t Make A Dumb Probability).
Theory About Issues” • Choose the highest EMV.
• CLEAR (Defining problem) Expected Value of Perfect Information (EVPI):
• RSU + VP (Model) EVPI = EV with Perfect Information – Maximum EMV.
• GIGO + R-D-I-M-S (Input Data sources) → Max amount worth paying for additional info.
• STEA (Solution methods) Expected Opportunity Loss (EOL):
• T-N-L-R (Testing) • Cost of not choosing best solution.
• SIS (Analyzing) • Always equals EVPI.
• HARD (Implementation problems)
• f ÷ (s – v) (BEP formula)
Sensitivity Analysis 2. Uncertainty → Outcomes known, but probabilities
• Examines how results change if probabilities or unknown.
inputs change. 3. Risk → Outcomes known, and probabilities
• Example: Thompson Lumber’s EMV changes if known.
probability of favorable market (P) shifts Decision Making Under Uncertainty
Decision Trees 1. MAXIMAX (OPTIMISTIC)
• Graphical version of decision tables. • Choose the alternative with the highest maximum
• Squares = Decision nodes payoff.
• Circles = State-of-nature nodes Example (Thompson Lumber):
• Steps in analysis: Define problem → Draw tree → • Large Plant: max = 200,000
Assign probabilities → Estimate payoffs → • Small Plant: max = 100,000
Compute EMVs. • Do nothing: max = 0
Decision = Large Plant (200,000)
Bayesian Analysis 2. MAXIMIN (PESSIMISTIC)
• Probabilities revised using new info. • Choose the alternative with the best “worst case.”
• Based on Bayes’ theorem. Example:
• Example: Survey results update probabilities of • Large Plant: min = –180,000
favorable/unfavorable market. • Small Plant: min = –20,000
• Do nothing: min = 0
Utility Theory Decision = Do Nothing (0)
• Utility = overall value of decision (not just money). 3. CRITERION OF REALISM (HURWICZ )
• People make decisions to maximize utility. • Weighted compromise between optimism (α) and
• Risk Preferences: pessimism (1–α).
o Risk avoider → prefers safety. Formula: Weighted Average = α(max in row) + (1–
o Risk seeker → enjoys risk. α)(min in row)
o Risk indifferent → neutral, straight-line Example (α = 0.8):
utility curve. • Large Plant = (0.8)(200,000) + (0.2)(–180,000) =
Example: Choosing between a safe bank deposit vs. risky 124,000
real estate investment. • Small Plant = (0.8)(100,000) + (0.2)(–20,000) =
76,000
• Do nothing = 0
Chapter 3: Decision Analysis — Expanded Reviewer Decision = Large Plant (124,000)
4. EQUALLY L IKELY (LAPLACE)
Introduction • Average payoff for each alternative.
• Decision theory → Analytic and systematic Example:
approach to the study of decision making. • Large Plant = (200,000 + (–180,000)) ÷ 2 = 10,000
• Good decision → Based on logic, considers all • Small Plant = (100,000 + (–20,000)) ÷ 2 = 40,000
available data and alternatives, uses quantitative • Do nothing = 0
approach. Decision = Small Plant (40,000)
5. MINIMAX REGRET
Six Steps in Decision Making • Build opportunity loss (regret) table:
1. Clearly define the problem. Alternative Favorable Unfavorable Max Regret
2. List possible alternatives.
3. Identify possible outcomes (states of nature). Large Plant 0 180,000 180,000
4. List the payoff (profit) for each alternative- Small Plant 100,000 20,000 100,000
outcome combination. Do nothing 200,000 0 200,000
5. Select a mathematical decision theory model. Decision = Small Plant (100,000 regret = minimum)
6. Apply the model and make a decision.
Example: Thompson Lumber Company is considering Decision Making Under Risk
expanding by manufacturing backyard storage sheds. EXPECTED MONETARY VALUE (EMV)
Alternatives = Large Plant, Small Plant, or No Formula:
Development.

Types of Decision-Making Environments

1. Certainty → Consequences of each choice Example (probabilities: Favorable = 0.5, Unfavorable =
known. 0.5):
 • Large Plant = (200,000)(0.5) + (–180,000)(0.5) = • Real Estate = 10,000 (p=0.8) or 0 (p=0.2).
10,000 • U(5,000) = (0.8)(1) + (0.2)(0) = 0.8.
• Small Plant = (100,000)(0.5) + (–20,000)(0.5) = ➡ Jane is indifferent if p=0.8, meaning she prefers
40,000 bank if probability is less.
• Do nothing = (0)(0.5) + (0)(0.5) = 0
Decision = Small Plant (40,000) QUICK TAKEAWAYS
EXPECTED VALUE OF PERFECT INFORMATION (EVPI) • Uncertainty models: Maximax, Maximin,
Formula: Hurwicz, Laplace, Minimax regret.
• Risk models: EMV, EVPI, EOL.
• Decision trees: Graphical EMV computation.
Example: • Bayes: Updates probabilities with new info.
• EV with Perfect Information (EVwPI) = • Utility: Considers personal preferences for risk.
(0.5)(200,000) + (0.5)(0) = 100,000
• Maximum EMV = 40,000
• EVPI = 100,000 – 40,000 = 60,000
➡ Thompson should pay up to 60,000 for perfect info.
EXPECTED OPPORTUNITY L OSS (EOL)
• Cost of not picking the best alternative.
Example:
• Large Plant: (0.5)(0) + (0.5)(180,000) = 90,000
• Small Plant: (0.5)(100,000) + (0.5)(20,000) =
60,000
• Do Nothing: (0.5)(200,000) + (0.5)(0) = 100,000
Decision = Small Plant (60,000 = minimum EOL)

Decision Trees
Steps:
1. Define problem.
2. Draw decision tree.
3. Assign probabilities.
4. Estimate payoffs.
5. Compute EMVs.
Thompson’s Decision Tree Example:
• Node 1 (Large Plant): EMV = (0.5)(200,000) +
(0.5)(–180,000) = 10,000
• Node 2 (Small Plant): EMV = (0.5)(100,000) +
(0.5)(–20,000) = 40,000
• Node 3 (Do Nothing): EMV = 0
Decision = Small Plant (best EMV)

Bayesian Analysis
• Probabilities revised using Bayes’ Theorem.
Example: Given positive survey results:
• P(Favorable | Survey Positive) = 0.78
• P(Unfavorable | Survey Positive) = 0.22
(Posterior probabilities are based on survey accuracy +
prior probabilities).

Utility Theory
• Utility = overall value of decision, not just money.
• People aim to maximize utility.
Risk Preferences:
• Risk Avoider: Prefers safer option (curves upward
slowly).
• Risk Seeker: Enjoys risk (steeper curve).
• Risk Indifferent: Straight line.
Example (Jane Dickson):
• Bank Investment = 5,000 guaranteed.