← All problems

Bezdek’s plank-covering conjecture for square annuli

Let Q=[0,1]2Q=[0,1]^2, and let HH be the concentric square of side length 1/21/2 whose sides are parallel to those of QQ. If finitely many planks P1,…,PmP_1,\ldots,P_m cover the square annulus Q∖HQ\setminus H, then their widths satisfy ∑i=1mwidth⁡(Pi)≥1\sum_{i=1}^m \operatorname{width}(P_i)\ge 1.

References

Source: arXiv

Progress summary

Refreshed
Open · solution reported

A new unrefereed preprint claims to prove the square-annulus conjecture and extend it to cube annuli, but the claim has not been independently checked.

Bezdek posed the annulus plank-covering question in 2003. The square case asks whether covering the annulus obtained by removing a sufficiently small concentric square requires total plank width at least the square’s minimum width.

Known results

  • Bezdek, 2003: posed the general annulus problem and verified the square case for homothety factor ε=1−2/2\varepsilon=1-\sqrt{2}/2.
  • Zhang and Ding, 2008: proved the conjectured bound for parallelograms and found counterexamples for equilateral triangles.
  • Smurov, Bogataya, and Bogatyi, 2010: proved the corresponding cube result for δ≤1−1/n\delta\le 1-1/\sqrt{n}.
  • Ambrus, Corbo, Goto, Leitz, and Mawhinney, October 2026: extended the cube result slightly in dimensions 22 and 33, but not to the conjectured threshold.

October 30, 2026 claimed proof

Egor Bakaev and Alexander Polyanskii claim that every plank covering of the specified square annulus has total width at least one, and state a multidimensional extension for unit cubes with an axis-parallel box removed. This is a claimed resolution, not an independently verified result.

Current status (as of November 2026): the square-annulus conjecture has a claimed but unverified proof; earlier partial results remain established, and the claim’s correctness is unresolved.

Sources

Solutions 0

No solutions have been posted yet.