Stage Recherche - Le Havre, France - CESI

CESI

Entreprise vérifiée

Le Havre, France

il y a 2 semaines

Posté par:

Sophie Dupont

beBee Recruiter

Description

Title:
A Dynamic Multi-objective Model For Campus Timetable Scheduling

Scientific domains:
Operations Research, Data Science, Education Management

Key-Words:
University timetabling, Carpooling, multi-objective optimization, simulation, sensitivity analysis

Supervisor:

Mohamed Amin BENATIA (EC), Houda TLAHIG (EC)

Research Work

Abstract of the internship

This internship aims to develop a dynamic multi-objective model for optimizing timetabling schedules within a university campus so as to maximize ridesharing and carpooling between students.

The model will integrate simulation and sensitivity analysis to identify optimal allocation of students to on-campus vs distance courses as well as optimal timetabling of courses/classes to minimize environmental impacts, traffic congestion, and maximize carpooling usage while respecting training quality constraints.

Internship Project

The objective of the internship is to develop a mathematical optimization model and algorithm to solve a dynamic multi-objective university timetabling problem that maximizes ridesharing between students.

The model will integrate linear/nonlinear programming, simulation, and sensitivity analysis.

Scientific context

University timetabling has garnered significant attention in the field of scheduling due to its inherent complexity, encompassing various constraints that need to be addressed.

Numerous existing approaches predominantly center on devising single-objective solutions, aiming to streamline the scheduling process by focusing on one dimension of optimization.

However, as the demand for more comprehensive and nuanced solutions grows, there is a discernible gap in the literature concerning dynamic multi-objective models for university timetabling.

This internship endeavors to contribute to the existing body of knowledge by developing a dynamic multi
- objective model that not only optimizes traditional factors like time efficiency but also incorporates emerging considerations such as environmental sustainability and transportation efficiency.

By introducing a dynamic aspect to the model, it adapts to the evolving needs and constraints of the university scheduling environment.

To further contextualize this research, previous works in the domain of multi-objective optimization and timetabling can be referenced.

For instance, studies by Burke et al and Li et al delve into multi-objective optimization techniques for timetabling, providing valuable insights that can inform the development of the proposed model.

Additionally, the work of Smith and Jones on incorporating sustainability considerations in scheduling processes serves as a relevant foundation for integrating environmental concerns into the proposed model.

While existing works have made commendable strides, certain limitations persist within the current literature.

First, many of the prevalent single-objective optimization approaches tend to oversimplify the intricate nature of the scheduling problem by focusing solely on one dimension, often neglecting the interplay between various factors.

This limitation can result in suboptimal solutions that fail to address the multifaceted constraints inherent in university timetabling.

Moreover, a substantial portion of the existing literature lacks a dynamic perspective, with solutions often being static and unable to adapt to evolving scheduling requirements.

This limitation is particularly significant as universities experience fluctuations in student enrollments, course offerings, and faculty availability, necessitating a more responsive and adaptive approach to timetabling.

Additionally, the current body of work in university timetabling tends to underemphasize the integration of emerging considerations such as environmental sustainability and transportation efficiency.

While some recent efforts touch upon these aspects, the incorporation remains limited and often lacks a comprehensive approach.

Consequently, there is a need for research that systematically integrates these evolving factors into the optimization model, reflecting the increasing importance of sustainability in academic institutions.

Furthermore, the majority of existing studies often employ static and predetermined sets of constraints, which may not fully capture the dynamic nature of real
- world university scheduling scenarios. An enhanced model should account for uncertainties, unexpected events, and real-time changes, ensuring adaptability in the face of unforeseen challenges.

To address these limitations and advance the state-of-the-art in university timetabling, this internship seeks to develop a dynamic multi-objective model that not only surpasses the constraints of single
- objective approaches but also incorporates adaptability, responsiveness, and a more comprehensive consideration of emerging factors. By doing so, this research aims to contribute significantly to the ongoing discourse and pave the w