A Comparative Study of EGD Based Hyper Heuristic for ETP

Abstract

This paper makes a comparative study of the Extended Great Deluge (EGD) based hype heuristic (HH) for the university exam timetabling problem. Indeed, Hyper-Heuristic is not new in AI field. It is an emerged search technology to select or generate (new) low level heuristics for combinatorial optimization problems. In the general framework of HH, it has two main stages: heuristic selection and move acceptance. For the latter stage, most of Meta heuristic algorithms are used. EGD has been firstly proposed and also used in HH as our previous job. Based on the numerous well-known papers and our previous experience, now, we investigate again the proposed EGD based HH to make the analysis of its performance and comparable with other methods in the literature or not. As another contribution for more comparison, Simple Random, Variants of Great Deluge(GD):Non linear GD( NLGD) and Flex Deluge(FD) are also employed in EGD based HH by applying it on the ETP.