Dynamic filters and randomized drivers for the multi-start global optimization algorithm MSNLP

Authors:
Zsolt Ugray;Leon Lasdon;John C. Plummer;Michael Bussieck
Affiliations:
Management Information Systems Department, Jon M. Huntsman School of Business, Utah State University, Logan, UT, USA;Department of Information, Risk, and Operations Management, McCombs School of Business, The University of Texas at Austin, Austin, TX, USA;Department of CIS/QMST, McCoy College of Business Administration, Texas State University, San Marcos, TX, USA;GAMS Development Corporation, Washington, DC, USA
Venue:
Optimization Methods & Software - GLOBAL OPTIMIZATION
Year:
2009

Citing 6
Cited 2

Stochastic global optimization methods. part 1: clustering methods

Mathematical Programming: Series A and B
Stochastic global optimization methods. part 11: multi level methods

Mathematical Programming: Series A and B
Scatter Search: Methodology and Implementations in C

Scatter Search: Methodology and Implementations in C
A Numerical Evaluation of Several Stochastic Algorithms on Selected Continuous Global Optimization Test Problems

Journal of Global Optimization
Multistart algorithms for seeking feasibility

Computers and Operations Research
Scatter Search and Local NLP Solvers: A Multistart Framework for Global Optimization

INFORMS Journal on Computing

Global optimization

Optimization Methods & Software - GLOBAL OPTIMIZATION
Pseudo-Cut Strategies for Global Optimization

International Journal of Applied Metaheuristic Computing

Quantified Score

Hi-index	0.00

Visualization

Abstract

We present results of extensive computational tests of (i) comparing dynamic filters (first mentioned in an earlier publication addressing a feasibility seeking algorithm) with static filters and (ii) stochastic starting point generators ('drivers') for a multi-start global optimization algorithm called MSNLP (Multi-Start Non-Linear Programming). We show how the widely used NLP local solvers CONOPT and SNOPT compare when used in this context. Our computational tests utilize two large and diverse sets of test problems. Best known solutions to most of the problems are obtained competitively, within 30 solver calls, and the best solutions are often located in the first ten calls. The results show that the addition of dynamic filters and new global drivers can contribute to the increased reliability of the MSNLP algorithmic framework.