Project Lead:
Dr Ignacio Castineiras (Ignacio.Castineiras@mtu.ie)
Project Status
In Progress
Project Lead Home Department:
MTU Cork - Science and Informatics - Computer Science
Alignment to HEA and NFETLHE SATLE 2022-23 Themes
- Education for Sustainable Development (ESD);
- Digital Transformation in the Tertiary Sector;
Alignment to MTU Strategy Themes
- Learner Education & Experience;
- People & Community;
- Leading Regional Development;
Project Description
MTU comprises 39 departments and ∼150 programmes, serving ∼20k students. Each Head of Department (from now on HoD) allocates the delivery of host modules to lecturers. The Timetable Co-ordinator (from now on TC) then interacts with all HoDs to facilitate the scheduling of the classes of almost all university programmes, assigning each class a room and timeslot. This is a well-known problem called the ‘Academic Timetabling Problem’.
There are two types of constraints in the problem:
- ‘hard’ constraints which must be satisfied in a valid solution, e.g. room capacities, 1hr lunch breaks for students/staff.
- ‘soft’ constraints which ideally would be satisfied, e.g. balanced number of lecture/lab hours per day, sufficient breaks among sessions, avoiding single lecture/lab days, and large gaps between timetable items for cohorts. The quality of a given solution is defined based on the satisfaction of these constraints.
Unfortunately, the timetabling problem is also computationally very difficult. In practical terms, given the number of classes and rooms in MTU, the problem becomes effectively intractable. and shared across departments. Finally, basic if-then-else rule-based software solutions fail to assist the HoD/TC in both finding a solution to the problem and evaluating its quality. The timetabling problem has a direct impact on staff/student well-being and overall learning experience. Suboptimal solutions can directly impact aspects such as class attendance (e.g. cohorts with single lecture/lab days, or a large gap between first and second lecture/lab on a given day). Class attendance is a consistent issue each year amongst lecturers, even more so for cohorts with Continuous Assessment modules. Students will be surveyed about timetable aspects that decrease likelihood of attendance, to incorporate their input to the problem model.
Given this context, the goal of this project is to create a Student-Centric Timetable Assistant Tool, that will be useful to the HoDs/TC in a number of ways:
- Supporting the direct specification of some ‘hard’ and ‘soft’ constraints stated by staff/students;
- Leveraging an ‘intelligent’ search, finding near-optimal solutions in a reasonable amount of time;
- Creating a plan (a sequence of actions) for adapting an existing timetable allocation to the new, and improved, solution. The tool can further be used in a simulation environment to assess the impact of facility/resource additions/removals.
What is the anticipated impact of this project?
As stated before, the timetabling problem is important, as ‘hard’ constraints must be imposed to preclude certain allocations (e.g., a group of students being scheduled with lectures and labs from 11am to 3pm and, therefore, not having time for lunch). Also, some ‘soft’ constraints must be imposed to cater for the student’s preferences and needs regarding timetable (e.g., by having a balanced number of lecture and lab hours per day, and by having enough time to brief among sessions, students can reflect better on the material being taught).
The proposed Student-Centric Timetable Assistant Tool aims leveraging Combinatorial Optimisation algorithms to find valid allocations (satisfying the ‘hard’ constraints) while also accounting as much as possible for the student needs (maximising the number of ‘soft’ constraints satisfied). In doing so, the Student-Centric Timetable Assistant Tool has the potential to improve the overall learning experience of the students and their satisfaction and belonging to the university. A similar impact on overall teaching experience is expected for academic staff, with better engagement and attendance from student cohort. The approach can develop and contribute to whole of-institution practice, as it can be applied cross-programme and cross-campus.
What will the outputs of this project be?
The primary output of the work will be the prototype Student-Centric Timetable Assistant Tool. The tool will be designed to be run by an individual in a local machine, without interfering with any other application. The prototype will receive as input the current timetable allocation and a set of hard/soft constraints. The input files can also anonymise the data, codifying staff/student/module names, etc. as indexes. The prototype will output a new timetable allocation, together with a sequence of actions for adapting the current timetable allocation to the new allocation.
Secondly, with the prototype, there will be a writeup of the evaluation of the prototype and use cases tested for dissemination amongst all stakeholders. It is expected that the research performed, and algorithmic approach developed will be further disseminated via publication in relevant conferences/journals.
Finally, results of the survey/f2f meetings with the student focus groups on their preferences and needs regarding timetabling will be gathered. This information will then be collated into a single document for dissemination amongst the HoDs.
Team Members:
This project aims to design, implement, and evaluate a student-centric timetable assistant tool. The project will be carried out entirely by the following 3 team members:
- Ignacio Castineiras, Lecturer, Department of Computer Science, MTU Cork.
- Diarmuid Grimes, Lecturer, Department of Computer Science, MTU Cork.
- Cemalettin Ozturk, Lecturer, Department of Process, Energy and Transport Engineering, MTU Cork.
The student voice is essential for the design of the student-centric timetable assistant tool. Students will act as feedback providers to the project:
- An initial questionnaire will be sent to the entire MTU student community at the beginning of the project, to gather information on the students’ preferences and needs regarding timetabling.
- A focus group of Student Engagement Ambassadors will be established, for them to meet regularly with the 3 team members to provide input in the tool and further defined the ‘soft’ constraints it is able to capture.
Likewise, for the tool to be applied, the project will also engage as feedback providers with the following line managers:
- Seán Mc Sweeney, Head of Department of Computer Science.
- Niall Morris, Head of the Department of Mechanical, Biomedical & Manufacturing Engineering.
- Ellen Crowley, Finance Manager.
- Clare O Leary, Timetabling Co-Ordinator.
- Irene Fenton, Head of IT Applications.
