مشخصات پژوهش

عنوان	Enumeration of Dominant Solutions: An Application in Transport Network Design
نوع پژوهش	مقاله چاپ شده
کلیدواژه‌ها	Enumeration; Dominant Solution; Branch-and-Bound Algorithm; Network Design Problem
چکیده	A One-Dimensional Binary Integer Programming Problem (1DB-IPP) is concerned with selecting a subset from a set of k items in budget constraint to optimize a goal function. In this problem a dominant solution is defined as a feasible selection to which no further item could be added in budget constraint. This paper presents a simple algorithm for Enumeration of Dominant Solutions (EDS) and investigates its functionality. The algorithm is then applied on one of the formulations of the Network Design Problem (NDP) as a case-study of 1DB-IPPs which arises in the transportation planning literature. The results are reported in detail for three illustrative examples and compared with the results of the Branch-and-Bound (B&B) algorithm. These examples suggest that in lower budget levels up to 40.2, 40.3 and 27.1 percentages the EDS algorithm outperforms the B&B algorithm. However, the overall performance of the B&B algorithm is notably faster in higher budget levels.
پژوهشگران	امیر علی زرین مهر (نفر اول)، یوسف شفاهی (نفر دوم)