Intro to Linear Programming
HTML-код
- Опубликовано: 30 июн 2024
- This optimization technique is so cool!!
Get Maple Learn ►www.maplesoft.com/products/le...
Get the free Maple Calculator for your phone►www.maplesoft.com/products/ma...
Play around with the exact Maple Learn document I use in the video ►learn.maplesoft.com/#/?docId=...
Play around with the Maple Learn document I use in the video ►learn.maplesoft.com/index.htm...
In this video we explore the idea of Linear Programming, which is an extremely powerful constrained optimization technique. It involves maximizing or minimizing a linear function with constraints a list of linear inequalities. The feasible region is all points satisfying those inequalities, and the big question is which points in the feasible region (which looks like a polygon) give the optimum values? The big idea of linear programming is that the optimal values occur at the vertices, that is where the iso-value line first touches the polygon.
0:00 Linear Programming
1:31 The Carpenter Problem
4:20 Graphing Inequalities with Maple Learn
5:45 Feasible Region
8:29 Computing the Maximum
10:36 Iso-value lines
13:15 The Big Idea
MY DIFFERENTIAL EQUATIONS PLAYLIST: ► • Ordinary Differential ...
Open Source (i.e free) ODE Textbook: ►web.uvic.ca/~tbazett/diffyqs
OTHER COURSE PLAYLISTS:
►DISCRETE MATH: • Discrete Math (Full Co...
►LINEAR ALGEBRA: • Linear Algebra (Full C...
►CALCULUS I: • Calculus I (Limits, De...
► CALCULUS II: • Calculus II (Integrati...
►MULTIVARIABLE CALCULUS (Calc III): • Calculus III: Multivar...
►VECTOR CALCULUS (Calc IV) • Calculus IV: Vector Ca...
►LAPLACE TRANSFORM: • How to solve ODEs with...
OTHER PLAYLISTS:
► Learning Math Series
• 5 Tips To Make Math Pr...
►Cool Math Series:
• Cool Math Series
BECOME A MEMBER:
►Join: / @drtrefor
MATH BOOKS & MERCH I LOVE:
► My Amazon Affiliate Shop: www.amazon.com/shop/treforbazett
SOCIALS:
►Twitter (math based): / treforbazett
►Instagram (photography based): / treforphotography
**TYPO** At 13:16 when I introduce the Big Idea I call the region concave when I mean convex!!!
I was wondering what a convex region would look like and I see this comment lol
Okay....I understood, thank you
Nice, i was just confused about that and see that now
Your enthusiasm is contagious and the way you presented the example, then the intuition and later the more formal geometric solution felt so much simpler than parsing the Wikipedia article. Thanks a lot!
It would be great to have a series of this topic. You would actually help a lot of not only math students, but those who are involved with economics, accountability, tourism, engineering, actuarial and computer sciences.
Great video, Dr. Trefor!
you channel is absolutely amazing, just wanna say i learn so much from watching it. thx for sharing your knowledge.
Glad you enjoy it!
Great explanation, and I can see you're passionate about this / about math, which is awesome!! Keep doing what you love and teaching with passion
Thank you Dr. Trefor, I was so confused in the lecture, your video is so nice and clear!
Amazing explanation! Just to point out that at 9:54 the actual value of f(0, 10) is not equal to 1800 but to 2000, having f(x, y) = 180x + 200y. Just a simple variable confusion. Thanks for this clear introduction to LP, Dr. Trefor.
Great explanation, you saved my studies today. Please, keep making videos
I hope professor Trefor Bazett could cover Convex Optimization in the future. Study with him is really energetic and engaging
A convex shape is one where each two points belonging to the shape can be connected with a straight line fully contained in the shape.
What a cool video! I'm introducing linear programming to my algebra students in 2 days and including a link to your nice video. I'm glad that I found your resources!
As cool as simplex is in concept, carrying it out is the most mind-numbing thing I've ever had to do in maths by miles
Haha that is true. But tbh when actually done in practice we're just to program it into the computer and get them to compute out the vertices.
@@DrTrefor yeah unfortunately A level further maths doesn't seem to appreciate that lmao. It doesn't go into stupid amounts of detail in the A level but I have had to do a two-stage simplex with 4 variables and 4 constraints in the past, which took me about 40 minutes to do the one question, it was pure suffering
@@vuraxis953 same for some uni courses. you have to do it manually
@@avanishparmessur5032 yup, I'm on the MMORS scheme at Cardiff now because I wanna go into stats and they have no maths and stats course without OR, and the OR modules do unfortunately have simplex in. Not looking forward to revisiting it
@@vuraxis953 interesting, im at cardiff too in data sci :)
this is amazingly simple in comparison to what i was looking for which is the actual simplex algorithm
Brilliant video. Thank you professor!
Glad you liked it!
Coming from an economics background this makes so much sense. I now know the math behind the concept of equilibrium 😄
This was really helpful. Thank you so much!
very comprehensive, thank you
Great explanation!
Great explanation. Please keep up the good work
Holy shit, thank you! Had to take one in my senior year anyways, might as well just preview for fun
I took a linear programming in uni years ago. I get a pass then that's it.
Now watching your video I truly know what it is about. Thanks.
This sounds more like graphical solutions of 2-decision variable LP problems. The simplex method requires conversion of the LP to standard form among other things I'm about to learn today in class. For those watching this and reading here, the cornerpoint method he shows is super easy. Find the x/y intercepts of each corner of the region, plug those (x,y) values into the objective function and find your MIN/MAX value from that table.
Great video nonetheless! Thank you
Wonderful explanation. Thank you for the video.
really great video on the concept
My teacher was talking about how we shift from vertices to vertices and also about some slack variables. Do you have a video for that, Sir?
Amazing! Nice explanation.
7:50 Actually all of the wood and all the labor does not always give one
the optimal solution. This depends on the slope of the optimization function. Thus one needs to check all the corner points, except for the origin.
In this case the corner points are: (0,10); (40/3, 10/3); (16,0)
If the Optimization function is:
a) 2y + 3x then the optimal point is (16, 0)
b) 3y + x then the optimal point is (0, 10)
Professor Charlie Obimbo
Yeah and you can’t have 10/3 tables
ILP should be used here
@@lukewitherow6380 Exactly!
U're the best. U just save me hours of head breaking maths
Free Great lesson, Thank you
Finally, a good video on the topic!
Just wanted to say you're a wonderful teacher
Thanks for the nice explanation.
This is really awesome....thanks!
LOVE THIS VIDEO💗💗💗
Loved your lecture and your T-shirt
Very clear in explanation.
7:00 4 vertices - due to 4 constraints
11:13 anhhh, the concept of *iso-line* is cool - i wanted some similar line/curve too when i was studying this chapter (Senior School) but didnt spend much time to think it out. But yeah, it makes many things much easier to communicate too.
So often this is taught purely algorithmically, but the geometric idea is so cool!
amazing, thank you!
Is there a video explaining for LP problems with >3 variables? The graph visualisation method would be extremely difficult with more variables. Thanks!
great job! nice teacher
nice vid!very informative
Great video sir! Are you planning to make more videos on linear programming?
Thank you! Yes, I do plan to! And move a bit more broadly into different optimization techniques (example discrete as well). However, I'm back to differential equations videos for the next few before I can do that.
Given any set of constraints are there algorithms that can calculate the optimal point efficiently using simplex geometry ?
Great Video. I had to Subscribe!
Thanks for the introduction! I find this fascinating, so have picked up Robert J. Vanderbei's book on Linear Programming. Though, I would love to know more from you as well - it would be a nice support to the material. Do you have any other videos, or are you planning to make some in the future? :)
I don't have more yet, but I plan to update this into a series at some point!
@@DrTrefor I hope you do more of these soon!!
You explained it far better than my college professors...16 years ago....
Any suggestions on where to find more videos on Linear Programming and the Simplex Method?
I attend Valdosta State University in Georgia. We have a course dedicated to going beyond this topic which is called Operations Research. The professor is encouraging of Data Science. We've covered this, slack variables, Tableau method, Anti-cycling rule, 2-Phase Simplex Algorithm for the 1st exam. Later we go on to learn MATLAB & R language.
Only in this case is the middle point maximum. It is very possible that the maximum actually lies on the axis. If the Isoprofit line has a steep enough slope max will be at y=0. If it's almost horizontal then max will be at x=0.
You really did justice to this topic in a brief time.
Thank you!
Thank you!!
Thanks a lot!!
I love your shirt 😂
thank you for the lesson :D
Great video ❤❤
Awesome!
The only linear programming tutorial that made sense🙌
Really nice presentation and great production. It appears that you refer to the region as concave, but I'm not sure that is correct. I believe it should be convex since the set of all feasible solutions should form a convex set. Additionally, I don't think you can guarantee solutions in the way you presented in a concave region of the plane.
Thank you! Preparing for a college course after 10 years of not doing math...I have 2 months to prepare haha, wish me luck!
A give the perfact examples for one to understand each and every bit of topic you introduce.🙏
Best explanation thanks 😊
You sir, are a lifesaver. Already saved me on multivariable calculus last semester, now saving me on optimization. Thank you!!
So glad I could help!
Tx for the lecture 🙏
You're most welcome!
I love you're T-shirt 😂
@Wilson Go
Yeah 😂 but believe it or not here in Iran we learn these things in highschool! I was so happy when I realized I don't need any college algebra course or precalculus when started to learn online.
*your
I’ve never seen it go the other way
Thank you sir... You are always there at the right time for me....🙂
I wanted lectures on Linear programming and fortunate that you have made lectures sir.... Thank you
Glad to hear that!
السلام عليكم.
اشكرك على الدرس.
Alsalaam Alikum.. Thank you.
How did he get the value of two Y???
20- 5 divide 4 times x
10- x divide 2
Thanks a lot. ❤️
You're welcome 😊
Hi, where did you get the figures from your solution at 8:58? thanks
I love this subject b/c it's so elegant and pretty simple. Is this vid a one-off or does it belong to a playlist?
Hoping to do a little series on optimization techniques, but for the next month or so it'll be a one-off as I head back to finishing off differential equations.
@@DrTrefor gotcha. I'll be sure to keep an eye for the other vids in as they appear. The ODE series is brill so glad to hear that you're putting your focus on that. It's been a great supplement to my self-study, so thank you!
@@DrTrefor it's already 9 months
And to think I racked my brain finding maximum and minimum values through differentiation.
Right?!?
Where the hell were you when I struggling with Calculus to the point that I gave up?
I came here to learn about the Simplex Method, but I stayed because of your amazing T-shirt
But why did u mention 'simplex' word here if it doesn't have any usage????
I just have to say, excellent, this video is excellent
Thank you!!
I WANT YOUR T SHIRT!!! I LOVE IT
Haha I love it so much:D
That intuition you present about why it should be the vertex not on one of the axis, is that always true? If the iso line had a different gradient, it would hit another vertex at its maximum, right? Or can it never have such a steep gradient?
The claim is it hits some vertex, so you have to check them all to see which it is
I have a question Dr. Brazzet why we need to construct a branch of knowledge i.e linear programming to deal with optimization problem when we have calculus methods like derivatives and Lagrange multiplier etc...
Love the shirt!
Excuse me sir, where did you get your tshirt from? I want it :)
i like the video! but how did you get x = 40/3 = 13.33 and y = 10/3 = 3.33 in the computing the maximum section? i'm having a hard time figuring it out
He took the two equations
5x + 4y = 80
10x + 20y = 200
and solved them using the substitution method, i.e. by making y the subject of the formula for each equation, and then equating the y's to eliminate the y's. And after solving for x (i.e. x=40/3), he got the value for y by using one of the equations which had y as the subject of the formula.
Prof. Charlie Obimbo
you are AMAZING
I was waiting for the point where you go back to acknowledge the nature of the problem space: the carpenter is not going to make any money for an unfinished item of furniture, so your model needs to allow only for integer values of x & y.
(FWIW my reason for looking up simplex method was because the news today in the UK was that school exam students will be given some extra information in advance about which topics will be in the exam papers; simplex method I remember as being the one topic in discrete maths that my whole class had trouble with, and eventually the teacher decided that it looked unlikely to appear in the exam. Unfortunately it did appear in that years paper… I feel like it may have been a different simplex algorithm that we covered, though).
The carpenter can finish the product in the next period, so if he can make 13.33 tables in 2 weeks, he can make 39 in 6 weeks. His optimization problem doesn't depend on integer values unless he is constrained to a short period.
The feasible region looks convex to me. Is that so?
Also, is the solution not integral values for x and y?
Oh good catch, yes absolutely convex not concave, I've pinned an explanation about that yes! Indeed this shows the precise location and then can round to the nearest integer depending on the context.
Lol happy that the moment when i will be doing this course on september i wont need to worry about the youtube teachers at least hahaha
are you ready bro? 😎
Hi this is a super nice video.
Can you make more videos on linear programming?
I want to do at least one more actually!
@@DrTrefor OK! only 1 more? linear programing is WAY more than just one concept.
How can I find the shirt you are wearing? I love it
Would you please do a video explaining steps for solving the simplex method, not using graphs
This analysis assumes someone values a third of a table and bookshelf equally to a full bookshelf and table. An additional constraint would be to consider only integer coordinates inside the feasibility region.
Love your shirt, where'd you get it from?
Can you suggest books that might have this type of problems?
I love the t-shirt! Where can I get one?
hi dr from where you got your t-shirt
Is this related to convex hull problems at all ?
Sir
please upload some content on the Fourier series with real-time Applications
Fourier series actually coming out in one week!
thank you..
Where did you get that awesome shirt?
thanks sir
love from India
9:27 Do x, y should be intgeres since we're talking about table/ shelves?
Yes, and we need to only accept integers for physical reasons we can find the "exact" answer and then search nearby to find the integer solutions, that works perfectly well.
that reveal at the 8:04 mark was exciting
this is a great explanation. to expound on the most money concept, you obviously wouldn't make money on 1/3 of a table or cabinet etc. How would you solve that so that the constraints are a whole number? Wouldn't that add another layer of feasibility and give a more accurate representation of money made?
That's a pretty funny shirt you got there.
From where can I get a shirt like yours?