[Defended thesis] Cyrille Ahmed Midingoyi: Semantic and modular representation of crop models using a declarative meta-language

Semantic and modular representation of crop models using a declarative meta-language

My name is Cyrille Ahmed Midingoyi. I'm a PhD student at INRA's Laboratoire d'Ecophysiologie des Plantes sous stress Environnementaux (LEPSE) in Montpellier. My subject aims to develop and implement a declarative language to represent plant growth models in a modular and semantic way. I have been a meteorological engineer at the Agence Nationale de la Météorologie du Bénin since 2009, where I was assigned to the Agrometeorology Department, then Climatology and finally Climate Research Department.
After my degree in meteorology, I obtained a master's degree in management computing, then a master's degree in geoinformation and in 2017 a specialized master's degree in Localized Information Systems for Territorial Planning at AgroParisTech (SILAT master's degree).
This thesis is not only a great opportunity for me to strengthen my skills and acquire new knowledge, but also an opportunity for the Research Department of the Benin Meteorological Agency.

• Starting date: January 2018
• University: MUSE Montpellier Université d’Excellence / Institut Agro
• PhD school: GAIA , Montpellier
• Scientific field:  IT– Agronomy
• Thesis management: Pierre Martre (Lepse, Inrae)
• Thesis supervisors: Pierre Martre (Lepse, Inrae), Frédérick Garcia (Miat, Inrae), Christophe Pradal (Agap, Inria)
• Funding: #DigitAg – Inrae
• #DigitAg : Cofunded PhD – Axe 4

Keywords: Declarative meta-language, plant growth models, semantics, modular

Abstract: Over the past two decades, the emergence of modeling platforms in agriculture has greatly increased the use of crop models in research, as well as their applications for production system management or scenario analysis. Despite the advances they represent, these platforms have impacted the models by causing a loss of transparency for the modelers, which has hampered the development of new formalisms, in particular for new uses related to phenotyping. There is thus a growing discrepancy between the representation of biological processes in crop models and our knowledge in plant ecophysiology. Here, we propose to design and implement a high level language, allowing the expression of each model and its composition independently of the programming languages and the formalisms of each modeling platform. This metalanguage will be based on the declarative languages developed in the international initiative COMBINE (‘COmputational Modeling in BIology’ Network; http://co.mbine.org/), which wi be extended to consider the specificities of agricultural models. The operational objectives concern the interoperability of the modeling tools and the links with the information systems collecting the data. Modularity of the models at the process level is aimed at allowing a better integration of knowledge on biological processes and facilitating the link with the data.

Jury compound:

Eric RAMAT, Professeur, Université du Littoral Côte d’Opale, HDR, Rapporteur
Gerhard BUCK-SORLIN, Professeur, AgroCampus Ouest, HDR, Rapporteur
Marianne HUCHARD, Professeur, Université de Montpellier, HDR, Examinatrice
Myriam ADAM, Chargé de recherche, CIRAD, Invitée
Gaëtan LOUARN, Chargé de recherche, INRAE, Examinateur
Pierre MARTRE, Directeur de recherche, INRAE, HDR, Directeur de thèse
Frédérick GARCIA, Directeur de recherche, INRAE, HDR, Co-directeur de thèse
Christophe PRADAL, Chargé de recherche, CIRAD, Co-encadrant

Contact :  
pierre.martre [AT] inrae.fr
frederick.garcia [AT] inrae.fr

Social networks: ResearchGate – LinkedIn

Communications & Papers

Upload the thesis manuscript

• Cyrille Ahmed Midingoyi, Christophe Pradal, Andreas Enders, Davide Fumagalli, Patrice Lecharpentier, Hélène Raynal, Marcello Donatelli, Davide Fanchini, Ioannis N. Athanasiadis, Cheryl Porter, Gerrit Hoogenboom, F.A.A. Oliveira, Dean Holzworth, Pierre Martre (2023), Crop modeling frameworks interoperability through bidirectional source  code transformation, Environmental Modelling & Software, https://doi.org/10.1016/j.envsoft.2023.105790