Discrete Global Optimization Problems with a Modified Discrete Filled Function

Abstract

This paper considers discrete global optimization problems. The traditional definition of the discrete filled function is modified in this paper. Based on the modified definition, a new discrete filled function is presented and an algorithm for discrete global optimization is developed from the discrete filled function. Numerical experiments reported in this paper on several test problems with up to 200 variables have demonstrated the efficiency of the algorithm.

Filled function method to optimize supply chain transportation costs

“…However, the filled function contains two parameters and the algorithm relies heavily on the selection of the parameters, it requires multiple adjustments in the iterative process to obtain the appropriate parameter values, which requires a large amount of calculation. In response to this shortcoming, later scholars made further improvements, and successively proposed filled functions [12,3,15,14,6,8,19,2,7,11,4,16,18,17,10,13,5]: The filled function proposed by paper [12] contains only one adjustable parameter, but the filled function is vulnerable to exponential terms. If the exponential term is too large, it is easy to cause memory overflow.…”