Continuous-time generalized fractional programming problems. II: an interval-type computational procedure (Q1949580): Difference between revisions

The paper is a continuation on the computational procedure for the continuous-time generalized fractional programming problem (CGFP) developed in Part I [the author, ibid. 157, No. 2, 365--399 (2013; Zbl 1285.90074)]. The author refines the discrete approximation method developed in Part I and extends the interval-type algorithm by the author [Taiwan. J. Math 16, No. 4, 1423--1452 (2012; Zbl 1286.90147)] to solve the problem (CGFP). The proposed computational procedure is a hybrid of a parametric method and a discretization approach. It is constructed a sequence of strictly decreasing upper and lower bound functions. The zeros of the upper and lower bound functions then determine a sequence of intervals shrinking to the optimal value of the problem (CGFP) as the size of discretization getting larger. By using these intervals, corresponding approximate solutions to the problem (CGFP) are found. The paper also establishes upper bounds of the lengths of these intervals, and thereby determines the size of the discretization in advance such that the accuracy of the corresponding approximate solutions can be controlled within the predefined error tolerance. Some numerical examples are provided.