Project description

Let's assume that a company has a truck that can ship goods. Each product that is loaded into the truck has a different potential profit whereas other products take up more space than others. The question is: how can we ensure that the truck is loaded with products that maximize the profit? In other words, which products shall the company load in the truck to get the maximum profit?

Possible solutions

The number of possible solutions for this problem are... over a million! For only 10 products, there are about 39.916.800 possible combinations! It is quite exhaustive to iterate all possible combinations and choose the best combination. What if we want to add more products in the future? Every complication will only increase the number of possible solutions.

In this project, we used genetic algorithms which seems to be a perfect fit for this kind of problems!

How it works

1. You choose the tab "Demostration" above
2. You define the problem. Specify which products exist in your problem and the capacity of the truck (e.g. 2 TVs of 0.400 cubic meters and a track of 3 cubic meters)
3. You click "run" and the "Genetic transportation of goods" will give you an optimal solution

Example

Let's assume that company A has a truck of capacity equal to 1m³. In their storage they have the following products available for shipping:

• Super 4K TV with margin=500$ and package space=0.2m³ X 5
• Suma shoes with margin=20$ and package space=0.05m³ X 10
• Notebook with margin=1000$ and package space=0.005m³ X 1

What would you choose in this situation? Obviously you wouldn't add the 5 TVs in the truck just to fill up the truck, right? You would most probably put the notebook first which doesn't take much space and yields the highest income. Then you would have space for 4 more TVs. You can add those too.

Afterwards, you are left with 0.15m³. So what do you do? You obviously add three pairs of shoes that will fill your truck and generate a few more bucks.

As simple as it sounds, it can be quite challenging to do this procedure when you have a warehouse of thousands of products. This is why the automation of such a procedure is needed. In this project, we used genetic algorithms to solve this issue. Click on the demostration tab to try such an example.