He-Yi Liu, Xinling li, Zhou Wei, Jen-Chih Yao, A modified outer approximation algorithm for convex MINLP problems
Full Text: PDF
DOI: 10.23952/jnva.11.2027.2.05
Volume 11, Issue 2, 1 April 2027, Pages 329-341
Abstract. This paper proposes a modified outer approximation (OA) algorithm for convex mixed-integer nonlinear programming (MINLP) problems. Building upon the conventional OA framework, we introduce an enhancement in the construction of the mixed-integer linear programming (MILP) master problem. Specifically, when the nonlinear programming subproblem is feasible, a new class of linear cuts is generated by incorporating an adjustable parameter into the gradients of the constraint functions. The theoretical validity of these linear cuts is rigorously proved. These cuts are subsequently integrated into the MILP master problem, forming the core of our modified OA algorithm. The effectiveness of the proposed algorithm is demonstrated through two numerical examples, where its performance is compared with that of the conventional OA method. The results demonstrate that the modified OA algorithm generates master problems with either a smaller scale or tighter constraints while maintaining a comparable number of iterations. This leads to a reduction in overall solution time and a significant enhancement in computational efficiency.
How to Cite this Article:
H.Y. Liu, X. li, Z. Wei, J.C. Yao, A modified outer approximation algorithm for convex MINLP problems, J. Nonlinear Var. Anal. 11 (2027), 329-341.
