Weighted Perimeters and pth Moments of Inertia of Convex Curves and Surfaces

Imagine a wire of fixed length, bent into a closed loop. If you were to spin this loop around a central point, how much resistance would it offer? This resistance, known in physics as the moment of inertia, depends entirely on how the wire is shaped and how far its various parts sit from the center of rotation. A loop that spreads its mass far out from the center is harder to spin than one that keeps its mass close in. For over a century, mathematicians have studied how to arrange such loops to either maximize or minimize this resistance. A famous result from the early 1900s showed that if you want to maximize this resistance with a specific type of weight, a perfect circle centered at the origin is the best shape. However, the question of what shape minimizes this resistance, especially when the wire must be convex—meaning it has no inward dents or sharp corners—has remained a puzzle. This is the territory explored by a new study that investigates how to arrange convex shapes to either maximize or minimize a weighted perimeter, a measure that combines the shape's boundary length with how far each point on that boundary sits from a central point.

The researchers, working in the abstract world of geometry, tackled a problem where the "weight" applied to the wire changes depending on its distance from the center. Sometimes the weight increases as you move away, and sometimes it decreases, becoming very heavy near the center. The team first proved that for almost any such weight function, in any number of dimensions, there is always a best possible shape that either minimizes or maximizes this weighted measure. They showed that these optimal shapes exist and are well-behaved, though they did not immediately reveal what those shapes look like. This existence proof is a crucial foundation, confirming that the mathematical search for a solution is not in vain.

The study then turned its attention to the specific case of two-dimensional space, where the shapes are flat curves. Here, the researchers focused on curves that possess a high degree of symmetry, specifically those that look the same when reflected across two perpendicular lines crossing at the center. They discovered a surprising truth about the shapes that maximize the weighted perimeter when the weight is a decreasing function, meaning the weight is strongest near the center. In this scenario, the optimal shape is not a circle or a square, but a "needle." This is a degenerate shape: a straight line segment that has been flattened so completely that it has no interior area, essentially collapsing into a line. The researchers proved that among all symmetric convex curves, this flat line segment is the unique winner for a wide family of weights, including those where the weight drops off as the distance from the center increases.

This finding challenges a previous intuition that the optimal shape might always be a solid object with an interior. The authors showed that for certain types of weights, specifically those that are highly singular near the center, the best strategy is to flatten the shape entirely. They also demonstrated that for other types of weights, such as those that increase with distance, the optimal shapes are solid and have a non-empty interior. The paper further established that for the specific case of minimizing the moment of inertia with a weight proportional to the square of the distance, the optimal shape is an equilateral triangle centered at the origin, confirming a result originally found decades ago but providing a new, simpler proof.

The team also explored what happens when the symmetry assumption is relaxed. They introduced a broader category of shapes called "four-quadrant convex," which allows for shapes that are not perfectly symmetric but still maintain a specific structure across the four quadrants of the plane. They found that the results regarding the needle shape holding true for the maximization problem still apply to these slightly less symmetric curves. The work relies on a sophisticated mathematical technique called majorization, which allows the researchers to compare different shapes by rearranging their parts in a specific order. By proving that a straight line segment is the best arrangement for a specific, simpler case, they were able to extend that logic to a vast range of more complex weight functions in one fell swoop.

Ultimately, the paper resolves a long-standing question about the geometry of convex curves under weighted constraints. It confirms that while solid shapes like circles or triangles are often the answer, there are specific conditions where the most efficient shape is a flat, one-dimensional line. The authors also identified a range of weights where the optimal shape is guaranteed to be solid, correcting earlier conjectures that suggested degeneracy might occur for all types of weights. Their work provides a complete picture of when a shape should be a solid object and when it should collapse into a line, offering a definitive answer to how convex bodies should be arranged to optimize their weighted boundaries.