Hi guys, I have learned MATLAB/SIMULINK during my uni studies, using it for a variety of applications such as DSP, communications, control systems, image processing and neural networks. I have also used it during my professional career after graduation.

I like to look at MATLAB projects from time to time to brush up my skills.

I came across what looks like a uni assignment. I have attached the PDF. But I didn't understand it at all. Not even what subject it was covering.

I tried googling some of the key terms but it was like reading a foreign language:

https://en.wikipedia.org/wiki/Barabási–Albert_model
https://en.wikipedia.org/wiki/Watts–Strogatz_model

Can someone give me a very high level overview of what subject this assignment is covering and what the point of this theory is for? What is the practical application of it?

I found uni to often give you a lot of complex theory without understanding the actual point of it, which for my learning style makes it far more difficult to learn.

I am always interested to learn about different applications and techniques that MATLAB can be used for, to expand my own capability and understanding.

Also what are the relevant MATLAB tool boxes and are there any other MATHWORKS features that are relevant?

And what is the sub-field called and where can I learn the basic theory?

If anyone can break this assignment down for me in very simple terms it would be much appreciated.

 

This is a brief and non comprehensive answer:

A graph is an abstract concept that is used to study and model real world systems. The "Traveling Salesman" is a typical problem, or the "The Seven Bridges of Königsberg" is another. These problems are used as examples of finding optimal paths for traversing a graph.

https://optimization.cbe.cornell.edu/index.php?title=Traveling_salesman_problem
https://www.mathsisfun.com/activity/seven-bridges-konigsberg.html

 

These types of problems show that a numerical solution is not always possible and, at some point, brut-force methods of trying every permutation cannot be analyzed so heuristic methods are used. Methods like genetic algorithms and statistical methods it is possible that tight, local optima are not found my these methods but optimal solutions on broader multidimensional surfaces can generally be found (or pretty good solutions) can be found if not an optimum. The problem is, the user does not know if tight local optimal solutions exist or not.

examples besides traveling salesman paths, scheduling a manufacturing plant that requires multiple manufacturing steps and makes multiple products with multiple production paths for some (or all) products to maximize annual production or minimize lead time or minimize inventory carrying costs (depending on the optimization criteria).
UPS and FedEx use similar tools every day for every route - with the added complexity of load-leveling (trying to make sure all trucks are working an equal amount of time to complete their routes).

another interesting problem I worked on was loading railcars at a multi-product manufacturing plant. Rail cars cannot pass each other on tracks and all products cannot be loaded at all railcar loading racks. If a railcar contained a product on the previous trip, it can be reloaded with the same product without cleaning - otherwise cleaning is needed. Oh, what a joy that was.