FSE100 - Analysis

Summary

This lecture delves into engineering analysis, defining it as a set of techniques to evaluate and understand system behavior. It distinguishes between model verification, ensuring a model meets specifications, and validation, confirming it aligns with customer expectations. The talk covers various analysis tools, including simulation, retrospective studies, and statistical methods, emphasizing their importance in iterative engineering design to refine models and meet requirements.

Takeaways

• 🔍 Analysis in engineering design involves breaking down an object or system to understand its basic building blocks and their relationships.
• 📏 Model verification is about ensuring the model behaves as expected by the engineers, focusing on internal mechanisms and specifications.
• 🌐 Model validation checks if the model meets the customer's expectations and performs as the system is supposed to, considering external inputs and outputs.
• 🔧 The iterative aspect of analysis is crucial for identifying and fixing issues in models, leading to improved and more accurate representations of the system.
• 🛠️ Various tools and techniques are available for analysis, including simulation, retrospective studies, statistical methods, and mathematical principles like calculus and linear algebra.
• 💡 The goal of analysis is to gain insights into the system's behavior, evaluate the model's quality, and ensure it meets both internal and external expectations.
• 🔄 The iterative process of engineering design involves cycling between modeling and analysis to produce a valid and verified solution that meets all requirements.
• 🧑‍💻 In software engineering, prototyping is a common technique to create a simplified version of the product for customer feedback and internal evaluation.
• 🔬 Statistical techniques are particularly useful for handling variability, a common challenge in engineering that can cause issues if not properly managed.
• 🔗 There's a close relationship between modeling techniques and analysis, with the latter often informing and improving the former throughout the engineering process.

Q & A

What is the primary focus of the lecture?

The lecture primarily focuses on the analysis from the perspective of engineering design, emphasizing the definition of analysis in engineering, the differences between model verification and validation, and the tools available for analysis.

How does engineering define analysis?

In engineering, analysis is defined as a set of techniques or a collection of tools used to evaluate a model or system, leading to an understanding of how the model or system behaves, allowing for better interaction with the environment and making changes to these models.

What is the difference between model verification and model validation?

Model verification is the process of ensuring that the model behaves as expected according to the internal specifications set by the engineers, while model validation is about ensuring that the model conforms to what the system is supposed to be doing and meets the customer's expectations.

Why is it important to perform both model verification and model validation?

Both model verification and model validation are important because they ensure that the model not only meets the internal specifications (verification) but also aligns with the external expectations and requirements of the customers (validation).

What is the iterative aspect of analysis within engineering design?

The iterative aspect of analysis within engineering design refers to the continuous process of improving the model by identifying issues through verification and validation, making improvements, and then re-analyzing the updated model to ensure it is valid and verified.

What are some common tools used for analysis in engineering?

Common tools for analysis in engineering include simulation, retrospective studies, statistical techniques, basic mathematical principles like calculus and linear algebra, and software prototyping.

How does simulation contribute to the analysis process?

Simulation contributes to the analysis process by allowing engineers to create a simulated environment to test 'what if' scenarios and evaluate whether the model performs as expected or as the customers expect it to.

What is the purpose of using retrospective studies in analysis?

The purpose of using retrospective studies in analysis is to collect past data, apply it to the current model, and evaluate how well the model performs relative to historical outcomes, with the goal of improving the system.

Why are statistical techniques important in engineering analysis?

Statistical techniques are important in engineering analysis because they help classify, identify, and manipulate variability, which can cause problems in engineering. By understanding and controlling variability, engineers can improve their models.

How do basic mathematical principles aid in engineering analysis?

Basic mathematical principles such as calculus, differential equations, linear algebra, and real analysis aid in engineering analysis by providing the necessary tools to find optimal points, solve systems of equations, and perform other mathematical operations that are crucial for model evaluation and improvement.

What is the role of software prototyping in the analysis process?

Software prototyping plays a role in the analysis process by allowing engineers to create a simplified version of the product that can be tested and evaluated by customers. This helps in gaining insights into whether the prototype meets expectations and functions as intended, guiding further improvements.

Why is the iterative nature of engineering important for analysis?

The iterative nature of engineering is important for analysis because it allows for continuous improvement of the model through cycles of modeling, analysis, identification of issues, and refinement. This iterative process ensures that the final solution is bug-free, meets customer requirements, and is both valid and verified.

Outlines

00:00

🔍 Introduction to Engineering Analysis

This paragraph introduces the concept of analysis in the context of engineering design. It defines analysis from an engineering perspective as breaking down an object or system to understand its basic building blocks and their relationships. The lecture aims to differentiate between model verification and validation, discuss available analysis tools, and emphasize the iterative nature of analysis in engineering design. The importance of gaining insights into a system's behavior and evaluating the quality of models is highlighted, with a focus on ensuring models conform to system expectations and customer needs.

05:01

🛠 Tools and Techniques in Engineering Analysis

The second paragraph delves into various common analysis tools and techniques used in engineering. It discusses simulation, both computer-based and physical, as a method to test models in simulated environments. The paragraph also mentions retrospective studies for analyzing past data against current models, statistical techniques for handling variability, and the application of mathematical principles like calculus and linear algebra. Software prototyping is introduced as a way to create simplified versions of products for customer feedback. The paragraph concludes by emphasizing the iterative process of engineering design, where analysis is used to identify and improve issues in models to meet specifications and customer requirements.

Keywords

💡Analysis

Analysis in the context of the video refers to the process of evaluating a model or system using various techniques to understand its behavior and performance. It is integral to engineering design as it helps engineers gain insights into the system's operation and identify areas for improvement. The video emphasizes that analysis is not just about breaking down an object but also about understanding the relationships between its basic building blocks.

💡Engineering Design

Engineering design is the overarching process in which analysis plays a crucial role. It involves creating, testing, and refining models to ensure they meet the requirements and expectations of both the engineers and the end-users. The video discusses how analysis within engineering design is iterative, aiming to improve models through continuous evaluation and refinement.

💡Model Verification

Model verification is the process of ensuring that a model behaves as expected according to the specifications set by the engineers. It is an internal check to confirm that the model's mechanisms are correctly implemented. The video uses the example of processing checks in a banking system to illustrate how verification ensures the model performs the intended functions correctly.

💡Model Validation

Model validation is the process of confirming that a model accurately represents the real-world system it is designed to emulate. It is an external check to ensure the model meets the customer's needs and expectations. The video contrasts validation with verification, highlighting that validation is more concerned with the relationship between input and output, rather than the internal mechanisms.

💡Iterative Process

The iterative process is a key aspect of engineering design and analysis, where models are continually refined through cycles of evaluation and improvement. The video stresses the importance of this process in achieving a valid and verified model that meets all specifications and performs as expected.

💡Simulation

Simulation is a common analysis tool mentioned in the video, which involves creating a virtual environment to test models under various conditions. It allows engineers to evaluate the model's performance in 'what if' scenarios and assess whether it meets the desired outcomes.

💡Retrospective Study

A retrospective study, also known as testing with unknown data, is an analysis method where historical data is applied to a model to see how it performs relative to past events. This technique helps in understanding how well the model would have worked in real-world situations before it was implemented.

💡Statistical Techniques

Statistical techniques are used in analysis to deal with variability and to classify, identify, and potentially improve the model's performance. The video mentions that variability can cause problems in engineering, and statistical methods provide a way to manage and optimize it.

💡Mathematical Principles

Mathematical principles such as calculus, differential equations, and linear algebra are fundamental to engineering analysis. These principles are used to solve complex problems, find optimal points, and ensure the model's integrity. The video highlights their importance in the analysis process.

💡Software Prototyping

Software prototyping is a technique used in software engineering where a simplified version of the product is created to gather feedback and insights. The video explains that prototyping allows for the early identification of issues and helps in refining the model before the final product is developed.

💡Customer Requirements

Customer requirements are the specifications and expectations that a model must meet to satisfy the end-users. The video emphasizes the importance of aligning the model with these requirements through validation to ensure the model's success in the market.

Highlights

Analysis in engineering design involves breaking down an object or system to understand its basic building blocks and their relationships.

Wikipedia defines analysis as applying scientific principles to reveal properties and the state of a system.

Engineering analysis is a set of techniques and tools used to evaluate a model or system for better understanding and interaction.

The purpose of analysis in engineering is to gain insight into the system and evaluate the quality of the model.

Model validation checks if the model behaves as the system is supposed to, ensuring it meets customer expectations.

Model verification ensures the model meets the specifications set by the engineers during the design process.

The difference between validation and verification is that validation is external, focusing on input-output relationships, while verification is internal, focusing on mechanisms.

Analysis tools in engineering include simulation, retrospective studies, statistical techniques, and mathematical principles.

Simulation allows for testing 'what if' scenarios and evaluating model performance in a simulated environment.

Retrospective studies involve applying past data to current models to see how they perform relative to historical outcomes.

Statistical techniques help in classifying, identifying, and improving variability within a system.

Basic mathematical principles like calculus, differential equations, and linear algebra are crucial for engineering analysis.

Software prototyping is a common technique in software engineering, creating a simplified version of the product for customer feedback.

The iterative nature of engineering involves cycling between modeling and analysis to improve the model.

The goal of analysis is to produce a model that is valid, verified, and meets all customer requirements by the end of the engineering design process.

Analysis techniques are closely tied to modeling techniques, and understanding one improves the other.

The iterative process of engineering design allows for the identification and improvement of issues in the model.

Transcripts

so in this lecture we are going to talk

about analysis from the perspective of

the engineering design more specifically

we're first going to define analysis

from the perspective of engineering and

then we're going to look at the

differences between model verification

and model validation we're gonna look at

some tools that are available to us for

analysis and then we're going to focus

on the iterative aspect associated with

analysis within engineering design there

are many different types of analysis

that exist and depending on your

background your discipline may be where

you've worked in the past you're going

to hear different types of definitions

about analysis teach engineering talks

about analysis from the perspective of

breaking down an object dealing with the

system the problem and fundamentally

looking at it at its basic building

blocks to create the essential features

and their relationships to one another

Wikipedia which tends to simplify a lot

of definitions looks at it as an

application of our overall scientific

principles to reveal some properties and

the state of the system that our model

is actually trying to represent from

engineering we tend to look at it more

so as a set of techniques it's a

collection of the tools available to us

that we can use to evaluate a model or

system and it will eventually lead us to

some understanding and some better

insights about how the model or the

system behaves allowing us to better

interact with the environment and make

changes to these models the reason why

we tend to do analysis and engineering

is primarily to gain some insight about

the system and to evaluate the overall

quality of our model when we're gaining

insight what we're learning about is the

way the system is supposed to behaving

and we're checking to see whether

whether or not our model actually

comports to what the system is doing

this is generally called model

validation we need to make sure we are

gaining an understanding of the

individual building blocks of our system

so that our model represents what the

system is supposed to actually be doing

on the other side of analysis we can

actually evaluate the quality of our

model to make sure that our model is

behaving in such a way that it is

matching to what we have an expectation

of there is one side of the system that

is building the way it's supposed to

build and on the other side

we want to make sure that what we are

putting into the system is doing what we

expect it to do this is an opportunity

for us to find mistakes in our models

it's an opportunity for us to find small

little issues that we might not have

originally thought existed and so with

the evaluation side of things we can

begin constructing more complete models

that represent what we want it to do

while the validation side of it is

already taking to account what the

system is supposed to be doing these two

sections are very important from a

perspective of engineering analysis we

must make sure that the model is

conforming to what the system is

supposed to be doing another way of

looking at it is whether or not our

model is actually performing to meet

whatever the customers expect it to mean

so if you think about possibility of a

bank coming to you and asking for you to

redesign how they process checks from a

perspective of model validation we need

to make sure that our model actually

processes checks and that it functions

in a way that the bank expects it to

function the other side of this process

is the verification side this is where

we are trying to make sure that our

model does what we expect it to do we

have an internalized process of how to

process those checks we need to know how

the checks are going to be moving

through the system we need to know how

information is stored how the money is

managed how the money is moved and that

is the verification side of things we're

going to make sure that when we are

verifying a model that it meets the

specifications that we have put forward

as the engineers in the design

requirements the bank doesn't

necessarily care about how the checks

are processed they only care about that

they are processed this is the distinct

difference between model validation and

model verification model validation is

sometimes seen as an externalized

process where we're more worried about

input and its relationship to output or

as verification tends to be more

internalized or worried about the

mechanisms by which input is turned into

output that is why we have to look at

both of these from an analysis

standpoint because if we miss the

validation process then there's a good

chance we're not meeting our customers

needs if we miss the verification

process there's a chance our model will

not perform based on anything it might

just fall

the building blocks are bet more of this

will come up and we begin talking about

testing and implementation because the

testing and implementation sections are

centered around improving our models to

a point where they are valid and

verified while going through the entire

iterative process of engineering design

when it comes to the actual process of

analysis there are more tools than we

can probably talk about we have

different disciplines within CID Z

ourselves we have the different majors

outside of Susie we have all the other

engineering majors we need to make sure

that we're covering all different types

of analysis tools but there's just too

many to count

so I'm going to go over some of the more

common ones that you will see within our

department that we like to focus on

first simulation tends to be the most

common this is either through computer

techniques or through life techniques

the whole idea is that we can simulate

an environment we can test what if

situations and from all of that we can

actually evaluate whether or not our

model is doing what we expect it to do

or whether or not the model is doing

what our customers are expecting it to

do another known method is what is

called testing unknown data usually

called a retrospective study we go back

into the past we collect data we apply

that data to the current model we're

looking at and we see how the model

performs relative to what happened in

the past hopefully we have an

improvement the whole idea is to improve

a system we can also use statistical

techniques when we're dealing with

things like variability variability is a

difficult piece to deal with when

dealing with engineering because

variability has a tendency to cause

problems and in statistical techniques

we can generally classify the

variability identify it figure out ways

to manipulate it or to improve it

ultimately improving our model overall

we also can consider some of the basic

math principles that you guys have been

put together over time things like

calculus differential equations linear

algebra even higher levels of like

number theory real analysis they all

play a role in analysis techniques for

engineers if we need to find the maximum

point or the minimum point you're going

to have to use calculus if we need to

find whether or not a system of

equations has a solution you're looking

at linear algebra all of these different

tools that you've put together over the

years will eventually create a

well-rounded buck

of analysis techniques that you can use

at any point in time in the realm of

software a very common technique is

software prototyping it is generally

cumbersome to create a finalized product

in the middle of our engineering design

process so what we want to do is we want

to create a slimmed down version one

that doesn't have as much details bugs

and/or bells and whistles and so we want

to create this slim down version that

our customers can look at and to give us

some insight as to whether or not it's

working the way they expect to work and

whether or not you think internally it's

working in all reality there are many

many more we have a possibility of any

sort of consideration and it's generally

based on modeling the technique that you

use for analysis is tied to modeling

there's an intimate relationship between

the two of them so as you learn more

modeling techniques you will learn more

analysis techniques throughout the years

ultimately the goal of this is to

improve our model we're trying to

produce the best possible model by the

end of this engineering design process

so it is important that we take the

steps necessary to slowly improve our

model because we are gaining some

insight and some better understanding

about our analysis and about our system

we can actually improve our model the

only way we can actually improve it is

if we have some sort of mechanism of

identifying the issues this is the whole

point of analysis we can identify those

issues either system-wide or internally

through the verification and validation

processes and from that we can actually

determine whether or not our model meets

all the specifications the customers

have put forth meets our own

internalized specifications and doesn't

have any major bugs or any major issues

that we need to repair the goal once we

spot these issues is to improve on them

constructing a new advanced better model

and then we go back and we analyze it

again this is the iterative nature of

engineering and this is what allows us

to create solutions that are bug free

hopefully produce the solution that we

expect it to produce and meet all of our

customer requirements because of this

iterative nature there's a lot of

relationship going on between modeling

analysis and subsequently the other side

of simulation and prototyping but

fundamentally as we do

cycling between modeling analysis we can

produce a solid solution that is valid

and verified and meets all of the

requirements we expect it to me this is

the goal of analysis thank you for

watching