|
Abstracts
Constributed Talks
An Analysis of Space Allocation in British Universities
E.K. Burke, D. Varley, University of Nottingham, UK
With the introduction of modularity, increasing student numbers
and the continued expansion of university departments, space in
British Universities is becoming an increasingly precious commodity.
To address this, some institutions have tried to ensure efficient
space utilization by employing methods such as Space Charging.
However, there are no rigid guidelines on which methods should be
used for space allocation within HE institutions.
This presentation will describe the results of a questionnaire
on the subject of space allocation which was sent to the estate
managers of ninety six British Universities in October 1996.
The questionaire asked questions in three specific categories
intending to discover how the requirements of each university differ.
Firstly, universities were asked about the nature and size of their
space allocation problem: how many buildings, rooms, departments,
schools are involved and what difliculties are associated with
allocating space? Secondly, we asked about how the problem is solved
at their institution, whether a manual or automated system is used
and whether computers are used at any stage in the process. Lastly,
we asked what qualities are required by each university in an efficient
space allocation process.
Thirty eight out of ninety six universities (40%) replied to the survey
of which 14 (37%) were former polytechnics. The analysis will group all
the replies together for a generalised view of the problem. However,
in specific parts, such as the constraints questions, comparisons will
be made between old and new style universities to analyse the differences
in approach used and the variety in space requirements.
The results of the analysis will be published in a full paper and will
be used to form a basis for the creation of a generalised automated space
allocation system over the next three years.
The presentation will conclude by making some comments, based on the survey
replies, as to what sort of criteria a generalised automated space allocation
system must meet.
Historical Developments, Present Situation, and Future Perspectives on Sports Timetabling
J.A.M. Schreuder, University of Twente, Netherlands
The article starts with an overview of published articles and books viewed
through different characteristics. One of the most important issues is the
use of Home-Away Patterns and Basic Match Schedules.
Also the number of matches per round - n-Factor in graph theory - is valued.
Based on the NP- hardness of sports timetabling, a number of approaches
like integer programming, simulated annealing and constraint satisfaction
are used. The role of decision support systems will be more important in the
future covering all kind of management aspects beside the scheduling part.
Incorporating Constraint Logic and TABU Search in the
Quest for the Perfect Timetable
G. White, J. Zhang, University of Ottawa, Canada
A number of approaches have been explored in the casting of timetables for
academic institutions. Two of these, the approach of Constraint Logic and the
approach of the TABU search, have given a number of promising results. The approach
to be described here combines these two approaches, the constraint logic approach
providing the initial first stage of the TABU solution. Since a pure constraint
logic approach yields a flat cost space (it either succeeds or it doesn't), use
of the TABU search is one way of incorporating a hierarchy of constraints.
This hierarchy can incorporate individual requests or organizational requirements
by weighing them according to some criterion. This combined approach has been tested
with real data from an educational institution and numerical results will be
presented and discussed.
Timetabling: Optimisation Heuristics for Rostering Problems
with Highly Constrained Resources
G. Vanden Berghe, KaHo St. Lieven, Belgium
We developed global optimisation schemes for Plane, which arranges the Fostering of
nurses in hospitals. The quality of a solution is quantified with a cost function,
taking in account a huge number of constraints. In our search for a converging
algorithm we tried to find a satisfying initial solution, define an appropriate
neighbourhood for a solution, determine the optimal (regarding tiine and improvement of
the solution) move in the neighbourhood, tune the parameters, and so on We tried to find
an equilibrium between the number of iterations of the algorithm necessary to reach an
acceptable roster and the number of evaluations of the cost function necessary to
perform these iterations.
Invited Talks
Marketing Timetabling Tools
A. Corbett, Corbett Engineering Ltd, UK
No abstract available
Graph Colouring Applications in Timetabling
M. Carter, University of Toronto, Canada
Most timetabling problems axe equivalent to an underlying graph colouring problem
with a number of side constraints. In this presentation, a variety of timetabling
applications will be described, and we will illustrate how they can each be formulated
as a graph colouring problem. Examples will include examination timetabling,
class-teacher timetabling, course timetabling and classroom assignment. Although the
side constraints on the problems in practice preclude the direct use of colouring
algorithms, we will discuss how colouring can be useful in analysing and solving these
problems.
The Project of Comparative Research of Different Approaches
to Timetabling Problems
V. Bardadym, International Renaissance Foundation,Ukraine
This lecture will be a declaration of a possible future international distant research
project. I have structured a research scheme, plan, criteria, etc. The timetabling
problems are considered as model problems for computer-aided management tools with
non-trivial problem solving components.
|
