Definition of Management Science

Management Science (MS) is the systematic use of mathematical models,

statistics, probability, optimization, simulation, and decision-analysis techniques
to help managers make better decisions.

It converts a real management problem into a quantitative model, analyzes

alternative solutions, and recommends the best feasible action considering
objectives, constraints, uncertainty, and available resources. Management
science is closely related to Operations Research (OR), which applies
scientific and mathematical methods to improve decisions in business and
operations. Standard operations-management texts commonly connect
management science with forecasting, quality, inventory, aggregate planning,
scheduling, maintenance, project management, and transportation models. [1][2]

General Management Science Process

1. Identify the problem – Determine what decision must be made.

2. Define the objective – For example, maximize profit, minimize cost, or

reduce waiting time.

3. Collect relevant data – Gather demand, costs, capacities, processing

times, and other information.

4. Develop a model – Represent the problem using equations, tables,

probability distributions, or simulation.

5. Solve the model – Use mathematical methods, spreadsheets, computer

software, or algorithms.

6. Validate the model – Compare the model’s results with actual business
conditions.

7. Implement the solution – Apply the recommended decision.

8. Monitor and improve – Review results and revise the model when
conditions change.

A management science model normally contains:

 Decision variables – Quantities that management can control.

 Objective function – The goal to maximize or minimize.

  Constraints – Limitations such as labor, material, budget, time, or
capacity.

 Parameters – Known data or assumptions.

 Uncertainty – Random demand, machine failures, delivery times, or

market conditions.

1. Introduction to Management Science

Management science supports rational decision-making in areas such as

production, logistics, finance, marketing, staffing, and supply-chain
management.

Main characteristics

 Uses quantitative and qualitative information.

 Focuses on the entire system rather than only one department.

 Examines trade-offs between cost, quality, time, capacity, and service.

 Uses models to compare possible courses of action.

 Helps managers make decisions under certainty, risk, and uncertainty.

Types of models

Model type Description Example

Physical A smaller or larger representation Factory layout

model of a real object

Analog model Uses a substitute representation Flow diagram

Mathematical Uses equations and symbols Linear programming

model

Simulation Imitates the behavior of a system Monte Carlo inventory

model over time simulation

Limitations
A model is only as reliable as its assumptions and data. It may simplify human
behavior, ignore qualitative factors, or produce an impractical solution if
managers fail to consider implementation issues.

2. Counting Sample Points

Counting sample points is used to determine the number of possible outcomes

in a probability experiment. The complete set of possible outcomes is called the
sample space, usually represented by S.

Basic counting rules

Multiplication rule

If an experiment has n1 possible outcomes in the first stage, n2 in the second

stage, and so on, the total number of outcomes is:

n1 ×n 2 ×⋯ × nk

Example: If a product may have 3 colors, 2 sizes, and 4 packaging types:

3(2)(4)=24

possible product combinations exist.

Addition rule

If alternatives cannot occur at the same time, add their possible outcomes:

n1 +n 2+⋯+ nk

Factorial

n !=n( n−1)(n−2)⋯(1)

Factorials are used in arrangements and permutations.

Permutations

A permutation is an arrangement where order matters:

 n!
P(n , r )=
(n−r )!

Combinations

A combination is a selection where order does not matter:

()
n!
C (n , r )= n =
r r ! (n−r)!

Application

Counting methods are useful in production sequencing, employee assignment,

product configuration, quality sampling, and decision analysis.

3. Introduction to Probability

Probability measures the likelihood that an event will occur. Its value ranges
from 0 to 1:

0 ≤ P( A)≤ 1

 P( A)=0: Event A is impossible.

 P( A)=1 : Event A is certain.

 A value close to 1 indicates a high likelihood.

Important terms

 Experiment – A process that produces an outcome.

 Sample space S – All possible outcomes.

 Sample point – A single outcome.

 Event A – One or more sample points.

 Complement Ac – Outcomes in which A does not occur.

Basic probability rules

Complement rule
c
P( A )=1−P ( A )
Addition rule

P( A ∪ B)=P( A)+ P( B)−P( A ∩ B)

If A and B are mutually exclusive:

P( A ∪ B)=P( A)+ P(B)

Conditional probability

P( A ∩ B)
P( A∨B)=
P( B)

This is the probability of A given that B has occurred.

Multiplication rule

P( A ∩ B)=P( A∨B) P( B)

If A and B are independent:

P( A ∩ B)=P( A) P(B)

Expected value

E( X )=∑ xi P(x i)

Expected value is the long-run average result of a random variable.

Management applications

Probability is used to estimate:

 Defective products.

 Customer demand.

 Machine breakdowns.

 Supplier delays.

 Project completion times.

 Investment and business risks.

4. Managing Quality

Quality is the ability of a product or service to meet customer requirements

and specifications. Quality management includes the activities used to prevent
defects, control processes, and continuously improve performance. [3]

Dimensions of quality

 Performance – How well the product performs.

 Features – Additional characteristics.

 Reliability – Probability of performing without failure.

 Conformance – Compliance with specifications.

 Durability – Expected useful life.

 Serviceability – Ease and speed of repair.

 Aesthetics – Appearance and sensory appeal.

 Perceived quality – Customer’s overall impression.

Quality management principles

 Focus on customer requirements.

 Prevent defects rather than merely inspect them.

 Involve employees in improvement.

 Use data to identify causes of problems.

 Improve processes continuously.

 Build quality into the design and production system.

Quality tools

 Check sheet.

 Pareto chart.

 Cause-and-effect diagram.

 Histogram.

 Scatter diagram.

 Flowchart.
  Control chart.

Cost of quality

Category Meaning

Prevention Cost of preventing

cost defects

Appraisal Cost of inspection

cost and testing

Internal Defects found

failure cost before delivery

External Defects found after

failure cost delivery

The goal is not simply to increase inspection. The better approach is to design
processes that consistently produce good outputs.

5. Inventory Management

Inventory consists of materials, components, work-in-process, finished goods,

and supplies held by an organization.

Reasons for holding inventory

 Meet customer demand.

 Protect against uncertain demand.

 Avoid production interruptions.

 Take advantage of quantity discounts.

 Separate different stages of production.

 Reduce the effect of supplier delays.

Costs of inventory

 Purchase cost – Cost of acquiring the item.

 Ordering cost – Cost of preparing and receiving an order.

  Holding cost – Storage, insurance, deterioration, and capital costs.

 Shortage cost – Lost sales, backorders, and customer dissatisfaction.

 Setup cost – Cost of preparing equipment for production.

Economic Order Quantity

The EOQ model determines an order quantity that minimizes ordering and
holding costs:

EOQ=
√ 2 DS
H

Where:

 D = annual demand.

 S = ordering or setup cost per order.

 H = annual holding cost per unit.

Reorder point

ROP=dL

Where:

 d = demand per period.

 L = lead time.

With safety stock:

ROP=dL+Safety Stock

Inventory control systems

 Continuous review system – Inventory is monitored continuously.

 Periodic review system – Inventory is reviewed at fixed intervals.

 ABC analysis – Classifies inventory according to importance or annual

value.

 Just-in-time system – Attempts to receive materials shortly before use.

 Safety stock – Extra inventory held to protect against uncertainty.

Inventory models are important in operations research because they balance
service requirements against ordering, holding, and shortage costs. [4]

6. Aggregate Planning and S&OP

Aggregate planning determines overall production, workforce, inventory, and

capacity levels for an intermediate planning period, usually several months to
one year.

The plan uses product families or aggregate units rather than individual
products. Forecasting is commonly the starting point for capacity, aggregate
production planning, and scheduling decisions. [3]

Sales and Operations Planning

Sales and Operations Planning (S&OP) aligns:

 Sales forecasts.

 Production capacity.

 Inventory.

 Workforce.

 Procurement.

 Financial objectives.

Main objectives

 Match supply with expected demand.

 Minimize total operating cost.

 Maintain acceptable customer service.

 Avoid excessive inventory.

 Use labor and equipment efficiently.

Planning alternatives

Adjusting supply

 Hiring or layoffs.
  Overtime or idle time.

 Subcontracting.

 Temporary workers.

 Inventory buildup.

 Backorders.

Adjusting demand

 Pricing changes.

 Promotions.

 Reservations.

 Delayed delivery.

 Product substitution.

Aggregate planning strategies

Strategy Description

Level Maintain a stable workforce and

strategy production rate

Chase Change production to follow demand

strategy

Mixed Combine inventory, overtime, hiring,

strategy and subcontracting

A typical aggregate-planning objective is:

[
Minimize total cost

Regular labor
+overtime
+hiring
+layoffs
+inventory
+shortage
+subcontracting
]

7. Short-Term Scheduling

Scheduling assigns jobs, workers, machines, and resources to specific times.

Short-term scheduling converts broader production plans into daily or weekly
operating decisions.

Scheduling objectives

 Minimize completion time.

 Minimize lateness.

 Reduce work-in-process inventory.

 Maximize machine utilization.

 Meet customer due dates.

 Reduce setup and changeover time.

Important terms

 Processing time – Time required to complete a job.

 Flow time – Time a job spends in the system.

 Due date – Required completion date.

 Lateness – Difference between completion time and due date.

 Tardiness – Positive lateness only.

 Makespan – Time required to complete all jobs.

Common priority rules

 FCFS – First come, first served.

 SPT – Shortest processing time.

 EDD – Earliest due date.

 LPT – Longest processing time.

 Critical ratio:
 Time remaining until due date
CR=
Work remaining

A critical ratio below 1 suggests that a job may be late.

Scheduling approaches

 Forward scheduling – Start as soon as possible.

 Backward scheduling – Work backward from the due date.

 Gantt chart – Displays jobs and activities across time.

 Finite-capacity scheduling – Does not assign more work than available

capacity.

8. Maintenance and Reliability

Maintenance includes activities performed to keep equipment operating or to

restore it after failure. Reliability is the probability that a system performs its
required function for a specified time under stated conditions.

Types of maintenance

 Corrective maintenance – Repair after failure.

 Preventive maintenance – Scheduled maintenance before failure.

 Predictive maintenance – Uses data and condition monitoring to predict

failure.

 Total Productive Maintenance (TPM) – Involves employees in

maintaining equipment and improving effectiveness.

Reliability measures

Reliability function

R(t )=P(T >t)

This is the probability that a component survives beyond time t .

Failure probability

F (t)=1−R(t)
Mean Time Between Failures

Operating time
MTBF=
Number of failures

Availability

[
Availability

\frac{MTBF}{MTBF+MTTR}
]

Where MTTR is mean time to repair.

Series and parallel systems

For a series system, every component must work:

R s=R1 R 2 ⋯ Rn

For two independent parallel components, the system works if at least one
works:

R p =1−(1−R1 )( 1−R2 )

Good maintenance improves safety, capacity, quality, delivery performance,

and equipment life.

9. Decision-Making Tools

Decision-making tools help managers select alternatives when outcomes may

be certain, risky, or uncertain.

Decision environments

 Certainty – The result of each alternative is known.

 Risk – Several outcomes are possible and probabilities are known.

 Uncertainty – Several outcomes are possible but probabilities are

unknown.

Decision table
A decision table lists alternatives, possible states of nature, payoffs, costs, and
probabilities.

Expected monetary value

EMV =∑ P(S i)×Payoff (S i)

Choose the alternative with the highest EMV when maximizing profit, or the
lowest expected cost when minimizing cost.

Expected opportunity loss

Opportunity loss is the difference between the best payoff for a state of nature
and the payoff from the selected alternative.

EOL=∑ P(S i)×Opportunity Loss

Decision tree

A decision tree shows:

 Decision nodes.

 Chance nodes.

 Possible outcomes.

 Probabilities.

 Payoffs.

Other tools

 Maximax – Selects the alternative with the best possible payoff.

 Maximin – Selects the alternative with the best worst-case payoff.

 Minimax regret – Minimizes the maximum possible regret.

 Sensitivity analysis – Examines how results change when assumptions

change.

 Value of perfect information – Measures the maximum amount worth

paying for complete information.
10. Forecasting

Forecasting is the process of estimating future demand, sales, costs,

workload, or other business conditions. Forecasting provides information for
capacity planning, inventory management, aggregate planning, and scheduling.
[3]

Forecasting approaches

Qualitative methods

Used when historical data are limited:

 Executive opinion.

 Sales-force estimates.

 Consumer surveys.

 Delphi method.

 Historical analogy.

Quantitative methods

Used when numerical historical data are available:

 Moving average.

 Weighted moving average.

 Exponential smoothing.

 Trend projection.

 Regression analysis.

 Seasonal models.

Moving average

A t + At −1+⋯+ A t−n +1
F t+ 1=
n

Where At is actual demand and n is the number of periods.

Exponential smoothing
 F t+ 1=α At +(1−α ) Ft

Where:

 F t+ 1 = forecast for the next period.

 At = actual demand.

 F t = previous forecast.

 α = smoothing constant between 0 and 1.

Forecast accuracy measures

Mean Absolute Deviation

MAD=∑ ∨ At −F t∨ ¿ ¿
n

Mean Squared Error

MSE=∑ ¿ ¿

Mean Absolute Percentage Error

MAPE=
100
n
∑ |
At −F t
At |
A forecast is not expected to be perfectly accurate. Its purpose is to reduce
uncertainty and support better planning decisions.

11. Productivity and Project Management

Productivity

Productivity measures how efficiently inputs are converted into outputs.

Output
Productivity =
Input

Types

 Labor productivity:
 Units produced
Labor hours

 Machine productivity:

Output
Machine hours

 Multifactor productivity:

Output
Labor + Materials+Energy +Capital

Productivity can improve through better technology, employee training, process

redesign, quality improvement, reduced downtime, and waste elimination.

Project management

Project management plans and controls temporary activities with a defined

objective, start date, and completion date.

Main project stages

1. Initiation.

2. Planning.

3. Execution.

4. Monitoring and control.

5. Closing.

Project network concepts

 Activity – A task that consumes time and resources.

 Predecessor – An activity that must occur first.

 Critical path – The longest path through a project network.

 Slack – Time an activity can be delayed without delaying the project.

PERT expected time

When optimistic, most likely, and pessimistic times are available:

 a+4 m+ b
t e=
6

Where:

 a = optimistic time.

 m = most likely time.

 b = pessimistic time.

Activities on the critical path require close monitoring because delays in them
delay the entire project.

12. Quality Function Deployment

Quality Function Deployment (QFD) translates customer requirements into

technical and operational requirements. Its main tool is the House of Quality.

Purpose of QFD

 Identify what customers value.

 Translate customer language into measurable specifications.

 Prioritize design characteristics.

 Improve communication among departments.

 Reduce design changes and development time.

Main parts of the House of Quality

1. Customer requirements – What customers want.

2. Importance ratings – How important each requirement is.

3. Technical characteristics – How the organization will satisfy

requirements.

4. Relationship matrix – Strength of the relationship between customer and

technical requirements.

5. Competitive assessment – Comparison with competitors.

6. Correlation roof – Shows positive or negative relationships among

technical characteristics.
 7. Technical priorities – Ranking of design requirements.

A common priority calculation is:

[
Technical priority

\sum(Customer importance)
(Relationship weight)
]

QFD helps prevent the organization from designing products based only on
engineering preferences rather than customer needs.

13. Monte Carlo Simulation

Monte Carlo simulation uses random numbers and probability distributions to

imitate the behavior of a system. It is useful when a problem is too complex for
a simple analytical solution. Operations-research texts include Monte Carlo and
discrete-event simulation as important tools for analyzing uncertain systems. [5]

General procedure

1. Define the problem and performance measure.

2. Identify uncertain variables.

3. Assign probability distributions.

4. Create cumulative probability intervals.

5. Generate random numbers.

6. Convert random numbers into simulated outcomes.

7. Repeat the experiment many times.

8. Analyze the results.

Example: demand simulation

Suppose demand probabilities are:

Dema Probabi Random-number

nd lity interval
 10 0.20 00–19

20 0.50 20–69

30 0.30 70–99

If the generated random number is 54, simulated demand is 20 units.

Important outputs

 Average demand.

 Probability of shortage.

 Expected profit.

 Average waiting time.

 Machine utilization.

 Inventory level.

Simulation does not guarantee an optimal solution. It estimates how a system is

likely to perform under specified assumptions, so the quality of the results
depends on the model, probability distributions, and number of replications.

14. Transportation Linear Programming Model

The transportation model determines how much product should be shipped

from each source to each destination at minimum cost while satisfying supply
and demand requirements.

Model components

 Sources or suppliers.

 Destinations or customers.

 Supply at each source.

 Demand at each destination.

 Unit transportation cost.

Mathematical formulation
Let:

 x ij = quantity shipped from source i to destination j .

 c ij = transportation cost per unit.

 si = supply at source i .

 d j = demand at destination j .

Objective:

Minimize Z=∑ ❑ ∑ ❑ c ij x ij
i j

Subject to supply constraints:

∑ ❑ x ij ≤ s i
j

Demand constraints:

∑ ❑ x ij ≥ d j
i

Non-negativity:

x ij ≥ 0

Balanced transportation problem

A problem is balanced when:

Total supply =Total demand

If supply and demand are unequal, a dummy source or dummy destination

is added.

Solution methods

 Northwest Corner Method.

 Least Cost Method.

 Vogel’s Approximation Method.

 Transportation simplex method.

  Spreadsheet Solver.

The transportation model is a special form of linear programming and is widely

used in distribution, logistics, warehouse allocation, and supply-chain planning.
Taha’s operations-research text specifically treats the transportation model as a
core linear-programming application.[6]

15. Linear Programming

Linear Programming (LP) is a mathematical technique for optimizing a linear

objective subject to linear constraints.

Components of an LP model

1. Decision variables – Unknown quantities to determine.

2. Objective function – Profit to maximize or cost to minimize.

3. Constraints – Resource, demand, capacity, or policy limitations.

4. Non-negativity restrictions – Variables cannot be negative unless

explicitly allowed.

General maximization model

Maximize Z =c 1 x 1+ c 2 x 2 +⋯+c n x n

Subject to:

a 11 x 1 +a 12 x 2+⋯+ a1 n x n ≤b 1

a 21 x 1+ a22 x 2 +⋯+a2 n x n ≤ b2

x 1 , x 2 ,… , x n ≥ 0

Assumptions of linear programming

 Proportionality – Contribution is directly proportional to the activity level.

 Additivity – Total contribution is the sum of individual contributions.

 Divisibility – Decision variables may take fractional values.

 Certainty – Coefficients are known and constant.

  Non-negativity – Activity levels cannot be negative.

Graphical method

For two decision variables:

1. Define the variables.

2. Formulate the objective and constraints.

3. Plot each constraint.

4. Identify the feasible region.

5. Find the corner points.

6. Evaluate the objective at each corner point.

7. Select the best feasible solution.

Simplex method

The simplex method systematically moves from one feasible corner point to
another until no further improvement is possible. It is useful for problems with
many variables and constraints.

Important LP terms

 Feasible solution – Satisfies all constraints.

 Optimal solution – Best feasible solution.

 Infeasible problem – Has no feasible solution.

 Unbounded solution – Objective can improve indefinitely.

 Slack – Unused amount of a resource.

 Surplus – Amount by which a ≥ constraint is exceeded.

 Sensitivity analysis – Examines how changes in coefficients affect the

solution.

 Shadow price – Improvement in the objective value from one additional

unit of a scarce resource.

LP is one of the central analytical tools in management science and operations

research, with applications in production planning, transportation, workforce
allocation, blending, scheduling, and resource allocation. [5][6]
Reference Guide

These notes synthesize the topics from the references supplied:

1. Martin, M. (2021). Production Management. Unlimited Books.

2. Morales, M. (2021). Business Organization, Operations & Management.

Mindshapers Co., Inc.

3. Martin, M. (2021). Production Management. Reader’s Knowledge

Bookstore.

4. Collier, Evans, and Lindsay (2020). Operations Management & Total Quality
Management. Reader’s Knowledge Bookstore.

5. Verma, A. P. (2019). Operations Management. International Books Royale,

Inc.

6. Hillier, F. S., and Lieberman, G. J. Introduction to Operations Research, 11th

ed.

7. Taha, H. A. Operations Research: An Introduction, 10th ed.

8. Powell, S. G., and Baker, K. R. Management Science: The Art of Modeling

with Spreadsheets, 5th ed.

9. Ravindran, A. R., Warsing, D. P., Jr., and Griffin, P. M. Supply Chain

Engineering: Models and Applications, 2nd ed.

[Link], D., and Tsitsiklis, J. N. Introduction to Linear Optimization.

The topics form a connected decision-making system: forecasting estimates

demand; aggregate planning sets overall capacity; inventory and
scheduling coordinate resources; quality and reliability protect
performance; and linear programming, decision analysis, and simulation
support the selection of effective alternatives.