Essential for Extended Supporting Hyperplane (ESH) method

Other Algorithms 1 implementation

Implementations

SHOT

Finds points on constraint boundaries for ESH hyperplane generation.

Purpose: Given interior point A and exterior point B, find point on boundary where constraint function equals zero (feasibility boundary).

Constraint Rootsearch:

  • findZero(ptA, ptB, ...): Search along line segment A→B
  • Returns (boundary point, exterior point) pair
  • lambdaTol: Tolerance on line parameter
  • constrTol: Tolerance on constraint violation

Objective Rootsearch:

  • findZero(pt, objLB, objUB, ...): Find objective level set
  • Used for objective function cuts

Algorithm:

  • Uses bisection or hybrid methods (TOMS 748)
  • Requires f(A) < 0 (feasible) and f(B) > 0 (infeasible)

Implementations: