This repository contains python and jupyter notebook files which can be opened in a Jupyter notebook environment (or viewed on Github.com), as well as executable python scripts which build graphs (which may take a while to load). For convenience, I have included saved stills of the graphs which can take a long time to create.
I defined this in the problem_function module which I import in all my metaheuristics code.
It plots this graph:
Where there is a local minimum (a very shallow one on this scale) at x = 0 and a global minimum at x = 101; so the best solution is 101.
The metaheuristics used were the following:
In a cloth factory there are 4 types of cloth to produce (A, B, C and D), producing a type of cloth requires a combination of different amounts of coloured wool and there is a finite amount of each wool.
| Wool colour | A | B | C | D | Wool available |
|---|---|---|---|---|---|
| Green | 1 | 2 | 1 | 1 | 10 |
| Red | 2 | 1 | 2 | 1 | 6 |
| Blue | 3 | 1 | 0 | 0 | 10 |
| Yellow | 1 | 4 | 0 | 0 | 18 |
| Brown | 0 | 0 | 1 | 3 | 8 |
| Purple | 0 | 0 | 3 | 3 | 12 |
The different cloths yield different amounts of profit
| Cloth | Profit |
|---|---|
| A | 3 |
| B | 5 |
| C | 4 |
| D | 1 |
These can be modelled into the following Linear Programming model:
The best solution I've found is A = 0.0, B = 4.5, C = 0.75 & D = 0.0 for a profit of 25.5.
The metaheuristics used were the following: