Decision Modelling: Introduction

Summary

This video introduces decision modelling, explaining it as a mathematical representation of scenarios to aid in business decision-making. It differentiates between deterministic models, where inputs are certain, and probabilistic models, where inputs are uncertain and estimated with probability. The video outlines the three main steps in decision modelling: formulation, where the problem and model are defined; solution, where the model is solved and tested; and interpretation, where sensitivity analysis is conducted to understand how changes affect outcomes. The summary emphasizes the importance of models in analyzing scenarios and making informed business decisions.

Takeaways

• 🏠 A model is a representation of the real thing, which can be physical or abstract, but in decision modelling, it's a mathematical or quantitative representation of a scenario.
• 🔢 Decision models are mathematical constructs that provide insights into solving decision problems, such as calculating total revenue as 6x where x is the number of units sold.
• 📊 Decision models are classified into two main types: Deterministic, where input values are known with certainty, and Probabilistic, where input values are uncertain and can only be estimated with some probability.
• 💼 Deterministic models are useful when dealing with known quantities like selling price per unit, number of parking spots, or store operating days.
• 🎰 Probabilistic models are essential for dealing with uncertainties such as customer purchase behavior, economic conditions, or government policies.
• 📈 The input values for models can be qualitative, describing non-numeric characteristics, or quantitative, involving numeric data like production hours or units sold.
• 🛠️ Decision Modelling involves three main steps: Formulation (defining the problem and developing a model), Solution (developing a solution and testing its correctness), and Interpretation (analyzing results and performing sensitivity analysis).
• 💡 Formulation involves defining the problem, such as determining profit, and developing a corresponding model like Total Revenue – Total Cost.
• 🔄 Solution involves acquiring input data, solving the model, and testing if the solution meets the objective, with the option to iterate and adjust the model based on the results.
• 🔍 Interpretation includes sensitivity analysis to understand how changes in input values affect the model's output, which is crucial for making informed decisions.
• 🔮 Sensitivity analysis helps in analyzing best- and worst-case scenarios and is useful for adjusting the model to accommodate new assumptions or making strategic decisions.

Q & A

What is a decision model?

A decision model is a mathematical or quantitative representation of a scenario that provides insights into solving decision problems a business might face.

What are the two main types of decision models?

The two main types of decision models are Deterministic models and Probabilistic (or Stochastic) models.

What distinguishes a deterministic model from a probabilistic model?

A deterministic model has input values that are known with certainty, while a probabilistic model has input values that are uncertain and can only be estimated with some probability.

Can you give examples of deterministic model inputs?

Examples of deterministic model inputs include a fixed selling price per unit, a specific number of parking spots for customers, the number of days a store is open, or a set warehouse space.

What are some examples of probabilistic model inputs?

Examples of probabilistic model inputs include the number of units customers might buy in a week, the possibility of an interest rate hike, or the potential for an economic recession.

What are the three main steps in decision modeling?

The three main steps in decision modeling are: Formulation, Solution, and Interpretation.

What is involved in the formulation step of decision modeling?

In the formulation step, the problem is defined and a mathematical model is developed to represent it. For example, a profit model might be created using total revenue minus total cost.

How is the solution step conducted in decision modeling?

In the solution step, the model is used to compute the desired outcomes, and these are tested to see if they meet the objectives. Adjustments can be made if the outcomes are not satisfactory.

What is sensitivity analysis in decision modeling?

Sensitivity analysis, also known as 'what-if' analysis, examines how the model responds to changes in input variables, helping to evaluate best- and worst-case scenarios.

Why might a business decide against maximizing short-term profits according to the script?

A business might choose not to maximize short-term profits to pursue long-term goals, adhere to company values, or comply with government regulations.

Outlines

00:00

🔢 Introduction to Decision Modelling

The paragraph introduces decision modelling as a quantitative approach to representing scenarios, using mathematical models. It compares models to physical replicas but focuses on their mathematical aspect. An example is given where a business's revenue is represented by the model 6x, where x is the number of units sold. The paragraph distinguishes between deterministic models, where input values are certain, and probabilistic models, where inputs are uncertain and can only be estimated with probability. It also differentiates between qualitative and quantitative data. The process of decision modelling is outlined in three steps: formulation, where the problem and model are defined; solution, where the model is solved and tested; and interpretation, which involves sensitivity analysis to understand how changes in input affect the model's output. The importance of sensitivity analysis in decision-making is emphasized, as it helps in considering best- and worst-case scenarios.

Keywords

💡Decision Modelling

Decision Modelling refers to the process of creating mathematical or quantitative models to represent scenarios and assist in making decisions. It is central to the video's theme as it outlines the framework for understanding and solving decision problems in a business context. The video discusses how these models can provide insights into business scenarios, such as determining profit or analyzing the impact of changes in selling price and production volume.

💡Deterministic Models

Deterministic models are a type of decision model where input values are known with certainty. The video explains that these models are useful when the variables, such as the selling price per unit or the number of parking spots, are fixed and can be precisely measured. An example from the script is knowing for sure that the selling price per unit is $6.

💡Probabilistic Models

Probabilistic models, also known as stochastic models, incorporate uncertain input values or data. The video emphasizes that these models are essential for dealing with variables that cannot be predicted with certainty, such as customer demand or economic conditions. An example mentioned is estimating how many units customers will buy in a week.

💡Qualitative Data

Qualitative data in the context of the video refers to non-numeric information that describes characteristics or qualities, such as customer attitude or government regulation. The video highlights that while qualitative data is important for understanding the context, it is the quantitative data that is often used in decision models to make numerical predictions and analyses.

💡Quantitative Data

Quantitative data is numeric information that can be measured and analyzed, such as the number of units sold or total revenue. The video uses quantitative data as the basis for the mathematical models that are central to decision modelling, allowing for calculations and predictions that can inform business decisions.

💡Formulation

Formulation is the first step in Decision Modelling, where the problem is defined and a model is developed to represent it mathematically. The video illustrates this with an example of defining a profit problem and developing a model that represents total revenue minus total cost.

💡Solution

Solution in the context of the video refers to the process of solving the mathematical model to find an answer to the defined problem. It involves using the model to calculate outcomes, such as determining profit by plugging in values for variables like the number of units sold.

💡Interpretation

Interpretation is the final step in Decision Modelling, where the results of the model are analyzed to understand their implications. The video describes how sensitivity analysis, a type of interpretation, can be used to see how changes in variables like production costs or sales volume affect the model's outcome.

💡Sensitivity Analysis

Sensitivity Analysis is a method of testing how different changes in input values affect the output of a model. The video explains its importance in decision modelling by showing how it can help analyze the impact of various 'what if' scenarios, such as reducing variable costs or increasing production.

💡Revenue

Revenue in the video is defined as the income generated from the sale of goods or services. It is a critical component of decision models, as it is used to calculate profit and is directly influenced by factors like selling price and the number of units sold.

💡Cost

Cost refers to the expenses incurred in the production and sale of goods or services. The video discusses two types of costs: fixed costs, which do not change with production volume, and variable costs, which vary with the number of units produced. Understanding and managing costs is essential for decision modelling in business.

Highlights

A model is a quantitative construct or mathematical representation of a scenario.

Decision models provide insights into solving decision problems in business.

Total revenue can be modeled mathematically as 6x, where x is the number of units sold.

Decision models are classified into deterministic and probabilistic models.

Deterministic models have input values known with certainty.

Probabilistic models have uncertain input values that are estimated with probability.

Input values can be qualitative, describing characteristics, or quantitative, providing numeric data.

The three main steps in Decision Modelling are Formulation, Solution, and Interpretation.

Formulation involves defining the problem and developing a model like Total Revenue – Total Cost.

Solution involves developing a solution and testing it against the model's objectives.

Interpretation includes analyzing results and conducting sensitivity analysis.

Sensitivity analysis helps understand how the model responds to changes in variables.

Business decision processes are often nonlinear due to multiple factors to consider.

A business may forgo short-term profit for long-term goals or due to company values and regulations.

Decision modelling is a systematic approach to aid in complex business decision-making.

Transcripts

This is a brief introduction to Decision modelling: You’ve probably seen a model house, a model

car, or a model plane.

In essence, a model is a representation of the real thing.

It could be physical or abstract.

But here, we are mainly interested in mathematical or quantitative models.

Therefore, in this context, a model is a quantitative construct or mathematical representation of

a scenario.

So, what does that mean?

Suppose a business sells its product at $6 per unit.

Then their total revenue can be written as 6x

where x is the number of units sold.

This is a mathematical model which we will refer to as a decision model because it can

provide insights into solving decision problems that the business might face.

Here is another model that shows the relationship between the list price of an item, the rate

of discount, and the net price a customer pays.

We will classify decision models into 2 main types:

Deterministic and Probabilistic models.

Deterministic means input values to are known with certainty.

For example, we can know for sure that our selling price per is $6 per unit.

Or that we have only 20 parking spots for customers, or that our store opens 6 days

a week, or that our warehouse space is 40,000 square feet.

On the other hand, probabilistic (or what we also call stochastic) models have uncertain

input values or data.

For example, how many units are customers going to buy this week?

Is the government going to raise key interest rate?

Is the economy going to be stable or will there be a recession next quarter?

In essence, the input values can only be estimated with some probability.

Now, these input values (or data) can either be qualitative or quantitative.

Qualitative, if they only describe characteristics or qualities that are usually non-numeric.

For example, customer attitude, government regulation, or store location.

And quantitative if they are numeric such as hours of production, units sold, and total

revenue.

Now, let’s examine the 3 main steps in Decision Modelling which are

Formulation, Solution, and Interpretation.

Under formulation, we first define the problem.

Suppose the problem is to determine profit.

Then we can develop a profit model, say, Total Revenue – Total Cost.

Suppose selling price per unit is $5, then total revenue is 5x.

Suppose fixed cost (FC) is $2,000 and variable cost (VC) is $2 per unit, then total cost

is 2000 + 2x.

Next, we acquire input data: Suppose our input value x is 500 units.

Then, under Solution, we’re going to develop a solution and test to see if it is correct

or if it meets our objective.

If we solve for profit here, we see that we have a loss of $500.

Since we don’t want a loss, we can retrace our steps back to the model to see where changes

can be made.

Let us tweak the model slightly by increasing the selling price per unit to $6, and change

the input data to produce, say, 800 units.

Then our profit will be $1,200.

If that meets the objective, then we can move on to interpretation.

This is where we analyze the results and do what is called sensitivity (or what if?) analysis

. Sensitivity simply refers to how the model

responds to changes.

That is, what if we reduce the variable cost per unit to $0.80 or increase production to

1,000 units or if fixed costs increase to $3000?

What will happen to profit?

So, sensitivity analysis is useful in analyzing the best- and worst-case scenarios in order

to make a good decision.

We can thus go back to the model to see if we need to make changes to accommodate the

new assumptions.

And if we like what we have overall, we can implement the solution.

In conclusion, while quantitative models are very systematic in nature, business decision

processes are usually nonlinear as there are typically many factors to consider.

A business may decide not to make the highest profit in the short term because of long-term

goals, or because of company’s values or government regulations.

And that’s a very brief intro to decision modelling.

Thanks for watching.