6.2.2. Case When n is Odd

So let us take the case where n is even. So remember, each edge has 2 endpoints. That means if I consider distinct edges of the graph, and if I focus on their endpoints, that will constitute the entire vertex set. So that means I cannot colour + 1 edges with the same colour, because if I do that, then their endpoints will give me n + 2 vertices, but my graph at the first place has only n vertices. So at most I can colour distinct edges with the same colour, I cannot colour more than those many edges.

This chunk discusses what happens in a graph when the number of vertices (n) is even. In a graph, each edge connects two vertices. If you try to use one colour on more edges than can be supported, you will exceed the total number of vertices since each edge uses up 2 vertex endpoints. Therefore, when n is even, you can at most use 1 less than half the vertices (or (n/2) - 1). So it's impossible to colour more edges with the same colour without exceeding the available vertices.

Think of a parking lot with a limited number of parking spots (vertices) and a certain number of cars (edges) that need to park. If there are too many cars trying to park in the same spots, some will be left without a spot. This is similar to how graph edges can't share vertex endpoints beyond a certain limit.

Whereas if n is odd, then this quantity is not well defined, it will not be an integer value. So in that context of an odd value of n, it will be easy to see that I cannot use a single colour to colour more than a certain number of distinct edges, because if I try to do that, say for instance, I tried to colour + 1 number of distinct edges, then their endpoints will give or constitute n + 1 nodes, but my given graph has only n nodes.

When n is odd, the reasoning changes slightly. The total number of endpoints for edges being colored doesn't reach a stable integer that can allow for more pairings of edges with vertex endpoints. This means that if you try to color more edges than allowed, you'll again exceed the total number of vertices available, hence leading to excess endpoints.

Imagine a group of 5 friends who want to pair up for sports—each sport requires a pair of friends. If they try to include an additional game with an unpaired friend, at least one sport won't have a complete pair. Hence, they can’t successfully have all their colors without leaving someone out.

Based on this information, I will try to solve question 11 where I want to find the edge chromatic number of a complete graph with n nodes and my solution will be divided into 2 cases depending upon whether my n is odd or even.

This chunk highlights the goal of calculating the edge chromatic number, which indicates the minimum number of colors needed to color the edges of a complete graph (where every pair of distinct vertices is connected by a unique edge). The analysis splits into two scenarios based on whether n (the number of vertices) is odd or even, acknowledging that these cases will yield different outcomes.

Think about a rainbow that can represent different colors of paint you have at your disposal. If you want to paint every line in a complex drawing (your complete graph), knowing the number of colors necessary depends on the complexity of the drawing—if the drawing can be fully paired up evenly, or if it leaves some lines hanging (like odd vertices).

I will show that these bounds on the edge chromatic numbers, which were the lower bounds because they were the least number of colours which are required, are actually tight in the sense I will give you a constructive colouring, a concrete colouring for colouring the edges of a complete graph with n nodes where n is even.

In this part, it is indicated that the bounds for the minimum number of colors needed are accurate—meaning you can't do it with fewer colors than calculated. A practical method or example of edge coloring for a complete graph with an even number of vertices will be provided, reinforcing the practical application of the theory discussed.

Imagine scheduling matches in a tournament where each team must face every other team exactly once. If there are an even number of teams, you can organize them in a way to efficiently use the minimum number of time slots (colors) without any overlap.

Now, let us see how exactly we can colour all the edges of a complete graph with n nodes where n is odd. And I will be using exactly n colours, which is optimal because the lower bound says that for n being odd, I need at least n colours.

In this segment, it discusses the coloring of a complete graph when n is odd, asserting that using n different colors is not just necessary but optimal. This is because each extra color corresponds to the need to compensate for the unpairable edges due to the odd number of vertices.

You can picture needing to set up game schedules for an odd number of teams in a sports league. To make sure every team plays against all others without overlapping games, you'll need to assign different time slots (colors), ensuring that every matchup is accounted for.

--