A Generalized Heron–Waist Problem: Optimality Conditions and Convergence Analysis

Manohar Choudhary and Triloki Nath and Ram K. Pandey · 2026

Read A Generalized Heron–Waist Problem: Optimality Conditions and Convergence Analysis by Manohar Choudhary and Triloki Nath and Ram K. Pandey (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-00071.txt.

Introduction

Choudhary, Nath and Pandey introduce and solve the Generalized Heron-Waist Problem, blending the classic Heron hub problem with the waist problem of least-perimeter layout. It seeks a closed weighted polygonal chain whose corners sit in given convex sets, while also cutting weighted distances to a central hub. The authors prove solutions exist under boundedness and general position, and are unique for strictly convex sets with positive weights. They derive optimality rules via subdifferential calculus, build a Projected Subgradient Algorithm, and prove it converges under shrinking step sizes.

Catalog record

CREATORManohar Choudhary and Triloki Nath and Ram K. Pandey
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE72334
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2603.00071

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

A Generalized Heron–Waist Problem: Optimality Conditions and Convergence Analysis Manohar Choudhary and Triloki Nath and Ram K. Pandey Abstract This paper introduces and solves the Generalized Heron–Waist Problem (GHWP), that integrates the classical Heron problem of optimal hub location and the waist problem of minimal-perimeter configuration. The GHWP seeks an optimal closed polygonal chain with weights whose vertices are constrained to lie in the given nonempty, closed, and convex sets, while simultaneously minimizing weighted distances to a central hub point. This coupled formulation naturally models systems in which cyclic internal connectivity and radial access to a hub must be optimized jointly—a structural feature that arises in applications such as supply-chain design, transportation planning, and communication infrastructures. Using modern convex analysis tools, we establish existence of optimal solutions under boundedness and general position assumptions of sets, we prove uniqueness when constraint sets are strictly convex with positive weights. We also derive first order necessary and sufficient optimality conditions using subdifferential calculus. For computation, we develop a Projected Subgradient Algorithm (PSA) and we prove convergence of the best-iterate sequence under classical diminishing step size rules. Numerical illustrations in ℝ 2 \mathbb{R}^{2} and ℝ 3 \mathbb{R}^{3} are provided to validate the algorithm’s robustness across diverse geometries and weighting schemes. Keywords: Convex Optimization, Heron Problem, Waist Problem, Facility Location, Subdifferential Calculus, Projected Subgradient Algorithm, Geometric Optimization. AMS Subject Classification: 51M04, 90C25, 90B85, 52A41, 65K05. 1 Introduction Geometric optimization problems, which seek to minimize distances, perimeters, or related quantities subject to spatial constraints, have fascinated mathematicians for centuries. One of the earliest and most celebrated distance-minimization problems is the classical Heron problem , attributed to Heron of Alexandria [ 5 , p. 304] . In its classical form, the problem asks the following: given two points P and Q located on the same side of a straight line \ell in the plane, find a point M on \ell that minimizes the sum of distances |M-P|+|M-Q| . Its elegant solution is governed by a geometric reflection principle, which reveals that the optimal point M satisfies an equal-angle condition : the line segments \overline{MP} and \overline{MQ} form equal angles with \ell , mirroring the law of reflection in optics. This foundational problem inspired numerous generalizations over subsequent centuries. In the seventeenth century, Pierre de Fermat posed what is now known as the Fermat–Torricelli problem : find a point in the plane that minimizes the sum of distances to three given points. Evangelista Torricelli provided its solution, establishing that when all angles of the triangle formed by the three points are less than 120^{\circ} , the minimizer lies in the interior of the triangle and forms 120^{\circ} angles with each pair of connecting segments [ 12 ] . Later, Endre Weiszfeld [ 13 , 14 ] extended this framework to n points by introducing an iterative algorithm that bears his name and remains influential in modern facility location theory. 1.1 Modern Generalizations via Convex Analysis The development of convex analysis in the mid-twentieth century, pioneered by Rockafellar [ 11 ] , Moreau, and others, provided powerful tools for revisiting classical geometric problems in a broadly applicable convex analysis framework. Motivated by applications in location science and optimization, Mordukhovich, Nam, and Salinas [ 8 ] introduced a convex-analysis generalization of the classical Heron problem by replacing points and lines with nonempty closed convex sets. In this framework, they formulated and solved the generalized Heron problem (GHP) : \min_{x\in S}D(x):=\sum_{i=1}^{n}d(x;C_{i}), (1) where S and C_{i} , i=1,\ldots,n , are nonempty closed convex subsets of \mathbb{R}^{s} , and d(x;C_{i}):=\inf\{\|x-y\|:y\in C_{i}\} denotes the Euclidean distance from the point x to the set C_{i} . Figure 1 : Illustration of the Generalized Heron Problem (GHP). This formulation replaces the classical constraint line with a closed convex feasible region S and points with convex target sets C_{i} , thereby substantially broadening the geometric and practical scope of the problem. Using tools from subdifferential calculus, the authors derived complete necessary and sufficient optimality conditions and developed convergent numerical algorithms based on projected subgradient methods. From an applied perspective, the generalized Heron problem models the search for an optimal hub location within a convex region that minimizes the total sum of distances to multiple convex sets. This development represents a fundamental transition of the Heron problem from classical geometric constructions to the modern theory of nonsmooth convex optimization. Recently, we further extended this problem to multiple hub sets [ 9 ] . Parallel to the Heron problem, another classical geometric optimization model is the waist problem , originally proposed by John Tyrrell [ 3 , p. 223] . The waist problem seeks a triangle of minimal perimeter whose vertices lie on three given, pairwise disjoint lines in space. Geometrically, it can be interpreted as an elastic band wrapped around rigid supports, contracting to a configuration of minimal total length. From an optimization perspective, the waist problem may also be viewed as a facility-location–type problem of distributed nature : rather than connecting facilities to a single hub, it optimizes the cyclic connectivity among the facilities themselves. Convexity plays a crucial role in this setting, ensuring the existence and uniqueness of solutions as well as equilibrium configurations. The optimal triangle satisfies reflection-type angle conditions that closely parallel those arising in the classical Heron problem, see [ 5 , p. 304] . Building on this classical foundation, We [ 10 ] recently introduced the generalized waist problem (GWP) , which extends the waist problem by replacing the lines with closed convex sets. The resulting optimization model is formulated as \min_{a\in C}D(a):=\sum_{i=1}^{m}\|a_{i}-a_{i+1}\|,\qquad a_{m+1}:=a_{1}, (2) where C:=C_{1}\times C_{2}\times\cdots\times C_{m}, with each C_{i}\subset\mathbb{R}^{n} being nonempty, closed, and convex. Here, the decision variable a=(a_{1},\ldots,a_{m})\in C represents a closed polygonal chain whose vertices are constrained to lie in their respective convex sets. This generalization significantly broadens the geometric scope of the waist problem and places it naturally within the modern convex-analysis optimization framework. Figure 2 : Illustration of the generalized waist problem in \mathbb{R}^{2} . Despite these advances, existing models continue to treat hub-based connectivity (as in Heron-type problems) and cyclic connectivity (as in waist-type problems) in isolation. In contrast, many real-world systems inherently require the simultaneous optimization of internal network connectivity and external access to a central facility . Typical examples include supply-chain loops connected to distribution hubs, transportation cycles linked to depots, sensor networks coordinated by a control unit, and communication backbones accessing centralized servers. Such systems cannot be adequately captured by either Heron-type or waist-type formulations alone. Motivated by this observation, we propose a hybrid geometric optimization framework that integrates both paradigms within a single model. Moreover, practical applications demand heterogeneity : different connections may lead to different costs, priorities, or levels of importance. To reflect this reality, we introduce heterogeneous positive weights that allow each interaction—whether between neighboring facilities or between facilities and a central hub—to be independently scaled, emphasized, or effectively neutralized. In this paper, we introduce and solve the Generalized Heron–Waist Problem (GHWP) , a unified convex optimization model that simultaneously captures hub-based and cyclic connectivity objectives under general convex constraints. In Section 1.2 , we formulate the GHWP and discuss its fundamental structural properties. Section 3 is devoted to the theoretical analysis of the problem, where we

Read the complete text