Setting the Research Agenda in Automated Timetabling: The Second International Timetabling Competition

Abstract

The 2nd International Timetabling Competition (ITC2007) was announced on the 1st August 2007. Building on the success of the first, this competition aimed to further develop interest in the area of educational timetabling while providing researchers with models of the problems faced which incorporate an increased number of real world constraints. A main objective of the competition was that conclusions drawn would further stimulate debate within the widening timetabling research community. The overall aim of the competition was to create better understanding between researchers and practitioners by allowing emerging techniques to be trialed and tested on real world models of timetabling problems. The competition was divided into three tracks to reflect the important variations which exist within educational timetabling within Higher Education. As these formulations incroporate an increased number of 'real world' issues, it is anticipated that the competition will set the research agenda within the field. After finishing on the 25th January 2008, final results of the competition are to be made available in May 2008. Along with background to the competition, the tracks are described here together with initial results for the datasets released.

Effective learning hyper-heuristics for the course timetabling problem

“…In order to perform a fair comparison, we followed the rules and instances of the Second International Timetabling Competition ITC-2007(McCollum et al, 2010, Track 2 (see Table 1). Competitors were required to obtain valid solutions (all hard constraints are satisfied) but there may be unplaced events (soft constraints).…”

“…The article contrasts two types of learning: static (offline) learning, in which there is a clear distinction between a training phase and an execution phase; and dynamic (online) learning, which takes place while the algorithm is solving a given instance. The resulting algorithms are tested over the set of publicly available real-world instances collected from the first and second International Timetabling competitions (McCollum et al, 2010). The next section formulates the course timetabling problem and describes the modeling methodology used.…”

Adaptive selection of heuristics for improving exam timetables

“…Only exams with common students and are four or less time slots apart are considered as violations -n is the number of students involved in the conflict -S is the total number of students in the problem Since then this problem has been used to test and compare many approaches in the literature. Recently, a more constrained set of benchmarks was made available as part of the International Timetabling Competition (ITC2007) [18]. The next section describes the ITC2007 dataset in detail.…”

“…This is due to the larger number of constraints it contains. A full description of the problem and the evaluation function can be found in [18]. In addition, the characteristics which define the instances are summarised in Table 2.…”

An investigation into two bin packing problems with ordering and orientation implications

“…Though simple to define, such "grouping" problems often pose significant challenges in practical settings because it is not always easy to judge whether the imposed constraints can be satisfied. Indeed many such problems including those concerning graph partitioning (Hertz et al, 2008;Isomoto et al, 1993;Jensen and Toft, 1994;Nakano1 et al, 1995), school and university timetabling (Lewis, 2008;McCollum et al, 2010), sports fixture scheduling (de Werra, 1988;Kendall et al, 2010;Rasmussen and Trick, 2008), load balancing (Falkenauer, 1998), and frequency assignment (Aardel et al, 2002;Valenzuela, 2001), are known to be NP-hard (Garey and Johnson, 1979;Karp, 1972), implying that we cannot hope to establish polynomially bounded algorithms for solving them in the general sense.…”