Go to contentGo to menuGo to searchGo to the news list

Laboratoire Angevin de Recherche en Ingénierie des Systèmes

Main navigation





    • Share this page on social networks
    • E-mail this page

      Send by mail

      Separated by coma
    • Print this page

    Research project MASCOT

    Medical Attendant Shift Conception and OpTimisation


    Team: Dynamic and discrete events and Optimization


    Labeling: none

    Term: 3 years (2018-2021)


    Funding: Atlanstic RFI 2020


    LARIS staff involved: Christelle Jussien-Guéret, David Baez


    Project partners: L. Billard (CHU Nantes), O. Bellenguez-Morineau (IMT Atlantique | LS2N)



    Abstract and objectives

    The objective of this project is to propose a new modeling and develop decision aid engines for the development of hospital staff schedules, in a context requiring to maintain continuity and safety of care in all circumstances, all by remaining attentive to the quality of life of caregivers.

    The traditional decision-making process for the construction of schedules is divided into several successive stages, each with different horizons: sizing of the teams in the year and establishment of work quotas (when part-time is granted), followed by the construction theoretical work cycles per service, and finally construction of nominative schedules in the month, amended daily in case of hazards. When the first phases have been carried out with too much approximation, the construction of monthly nominative schedules may become unrealizable and lead to an additional cost through the use of substitutions or overtime. The various optimization problems underlying this decision-making process are for the most part addressed in the literature. They are however studied in isolation from each other. In addition, the vast majority of  works for hospital's staff planning revolves around a goal of controlling the costs of the system. However, the current context of tension (highly constrained resources) leads to bring out first of all a need primarily centered on the feasibility, as well as the ability to absorb hazards. On the other hand, we note that human aspects, highlighted by the current suffering of caregivers, must be taken into account in order to allow staff to recover a certain quality of life at work.

    The proposed study therefore focuses on three major scientific obstacles: first, it will be a question of finding a way to integrate the various decision horizons in a same approach in order to allow a more relevant global optimization, in particular because the sizing of the teams has a direct influence on the schedules that can be built. In a second time, it appears necessary to provide robust solutions to manage daily hazards with greater flexibility, allowing easy changes schedules while maintaining their validity. Lastly, a fundamental and interdisciplinary reflection aimed at integrating human realities, beyond the quantifiable aspects usually tackled in Operational Research, will enrich the modeling of the problems treated in an innovative way.

    This project involves three partners: the LARIS laboratories and LS2N, whose one of specialties is the solving of optimization problems in logistics and production, and the Nantes University Hospital Center, which will bring its expertise on the issues addressed and provide data. The reflections will also be nourished by exchanges around the consideration of human resources with the associations ANACT (National Association for the Improvement of Working Conditions) and ARACT (Regional Association for the Improvement of Working Conditions).

    In the current process of overhauling the health system, we believe that the implementation of operational research approaches on real data is a guarantee of a better knowledge of the problems to be treated, as well as a sure source of increase in importance in the processing efficiency of these issues. The developed optimization engines as well as the experience acquired by the laboratories on these issues during this project will make it possible to consider later the development of generic models applicable to other health establishments.