A Graphical Approach to the Frobenius Number Problem in Three Variables

Xinxin Fang

2024-01-17

Abstract:The Frobenius number for a set of relatively prime positive integers, where the smallest integer in the set is at least 2, is the largest integer that cannot be expressed as a nonnegative linear combination of those integers. We analyze the Frobenius number for three variables by analyzing the lattice points associated with the line $ax+by = ab$ and its downward parallel translations, along with each lattice point's corresponding value $ax+by$. As an application of our graphical approach, we recover key theorems such as the results from Sylvester and Selmer by reducing them to a graphical problem of locating a lattice point, followed by an evaluation of the linear form $ax+by$ there. Therefore, an arithmetic problem of finding the Frobenius number is reduced to a visual/graphical problem of locating the lattice point, followed by an evaluation, both of which are simple tasks. We then derive relatively simple formulae for the general Frobenius number of three variables. Compared to known results, our results (Theorem 5.1) are not algorithmic in nature nor intricate, which constitutes a significant improvement.

Number Theory

What problem does this paper attempt to address?

### What problem does this paper attempt to solve?

This paper aims to solve the ternary Frobenius number problem. Specifically, given three relatively prime positive integers \(a, b, c\) (where the smallest integer is at least 2), the Frobenius number \(g(a, b, c)\) is defined as the largest integer that cannot be expressed as a non - negative linear combination of these integers. For the two - variable case, there are already explicit solutions, but for the three - or - more - variable case, there is currently no explicit solution.

### Main contributions of the paper

1. **Graphical method**: By analyzing lattice points related to the line \(ax + by=ab\) and its downward parallel translations and their corresponding linear values \(ax + by\), the author transforms the arithmetic problem into a graphical problem.

2. **Simplified formula derivation**: The author derives a relatively simple formula for the ternary Frobenius number. Compared with existing algorithms and complex results, this method is more intuitive and simple.

3. **Special case handling**: The author pays special attention to the case when \(c\) is in the form of \(ab - ax - by\) (where \(x, y>0\)) and defines these \(c\) as "exceptional values", so that \(g(a, b, c)<g(a, b)\).

4. **Verification of existing results**: As an application, the author re - obtains the results of Sylvester (1882) and Selmer (1977) through the graphical method, further verifying the effectiveness of the new method.

### Main conclusions

Finally, the author proposes a new formula applicable to the ternary Frobenius number:

\[g(a, b, c)\]

It is divided into the following cases according to different conditions:

1. When \(a\equiv0\pmod{l}\):
\[g(a, b, c)=ab - a((q - 1)h+ 1)-b\]

2. When \(a\equiv1\pmod{l}\):
- If \(b < ah\):
\[g(a, b, c)=ab - a((q - 1)h+ 1)-b(r + 1)\]
- If \(b>ah\):
\[g(a, b, c)=ab - a\left(\left\lfloor\frac{ah - h}{l}\right\rfloor+ 1\right)-b\]

3. When \(a\not\equiv0, 1\pmod{l}\):
- If \(b>\frac{ah}{r}\):
\[g(a, b, c)=ab - a(qh + 1)-b\]
- If \(b<\frac{ah}{r}\):
\[g(a, b, c)=ab - a((q - 1)h+ 1)-b(r + 1)\]

### Future work

The author plans to further study the Frobenius number formula for more than three variables and explore the possibility of extending the current method to more variables. In addition, they also propose some potential research directions, although these works are still in progress.

### Summary

This paper simplifies the solution process of the ternary Frobenius number problem by introducing a graphical method, providing a new and more intuitive way to understand and calculate the Frobenius number. This not only contributes to theoretical research but also provides an effective tool for practical applications.