Mohammed-Mahmoud Ould-Sidi Memmah
AgreenSkills session, year: 1st session, 2013
Receiving laboratory: CERCIA, University of Birmingham, United Kingdom
Country of origin : France
Meta-heuristics for the model-based design of integrated cropping systems
The model-based design of Integrated Cropping Systems (ICS) is an objective of my home research unit PSH. It aims to identify all combinations of a number of technical actions to be implemented to satisfy a set of nonlinear and conflicting objectives (environmental protection, economic viability, organoleptic quality, ecological restoration) subject to strong constraints. Therefore, the model-based design of ICS is studied as a multi-objective constrained optimization problem. This is a difficult optimization problem where there are no explicit analytical mathematical formulas of the objectives’ functions in terms of the decision variables. This class of optimization problems is thus intractable using classical optimization methods. Many bio-inspired meta-heuristics, e.g. Evolutionary Algorithms (EAs), ant colonies, the particle swarm optimization and artificial immune systems, have been proposed to tackle very hard optimization problems. These algorithms have been deeply used in experimental studies and have often produced efficiently good quality solutions. Unfortunately, the rare theoretical studies of these algorithms are mainly restricted to very simple forms of evolutionary algorithms. Theory is important because it gives us accurate and deeper understanding of why, how and when meta-heuristics work effectively. This information could lead to the design of new algorithms and their successful applications to new optimisation problems.
This is the focus of an ongoing project at CERIA studying theoretically what types of problems can be solved using what kind of EAs, and why. The relationship between problem characteristics and algorithmic parameters is particularly investigated. Five population-based meta-heuristics are studied and many optimization problems are considered as case studies. Three case studies that would be investigated during the stay relate the model-based design of ICS and deal with pesticide use reduction, landscape ecosystem services and sustainability, and fruit breeding.
My research focuses on the development of multi-objective optimization algorithms and their applications to many real-world problems. During my PhD at Centrale Lille supervised by Professor P. Borne, I was interested in the real-time regulation of urban collective transportation networks. As a postdoctoral researcher at Université du Littoral Côte d’Opale and GDF, I had worked on the optimal design and dimensioning of hydrogen distribution pipeline networks.
As an INRA researcher in the PSH unit, I am studying the model-based design of Integrated Cropping production Systems (ICS) as difficult multi-objective optimization problems. This approach interfaces models describing the cropping systems with meta-heuristics algorithms. The used models describe either the interactions between plants, pests, and natural enemies under the effect of cultural and land use practices, or the interactions between genotypes, environments and cultural practices. Our short-term objective is to propose efficient algorithms able to solve these kinds of optimization problems. Thus, we will propose effective cultural procedures and land use scenarios satisfying many antagonist criteria (economic, environmental, ecological, and agronomic) and fulfilling some biotic and abiotic constraints.
J. André, S. Auray, J. Brac, D. De Wolf, G. Maisonnier, M-M. Ould Sidi, A. Simonet. Design and dimensioning of Hydrogen transmission pipeline networks. European Journal of Operational Research 229 (2013), pp. 239-251. (alphabetical authors’ order)
I. Grechi; M-M Ould-Sidi; N. Hilgert; R. Senoussi; B. Sauphanor; F. Lescourret. Designing integrated management scenarios using simulation-based and multi-objective optimization: application to the peach tree-Myzus persicae aphid system. Ecological Modelling 246 (2012) 47–59.
B. Quilot-Turion, M-M. Ould Sidi, A. Kadrani, N. Hilgert, M. Génard, F. Lescourret. Optimization of genetic parameters of the ‘Virtual Fruit’ model to design peach ideotypes for sustainable production systems. . European Journal of Agronomy 42 (2012) 34– 48.
A. Kadrani, M-M. Ould Sidi, B. Quilot-Turion, M. Génard, F. Lescourret. Particle swarm optimization to design ideotypes for sustainable fruit production systems. International Journal of Swarm Intelligence Research 3(2), 1-19.
M-M. Ould-Sidi and F. Lescourret. Model-based design of innovative cropping systems: state of the art and new prospects. Agronomy for sustainable development. Volume 31, Issue 3 (2011), 571-588.
D. De Wolf, M-M. Ould Sidi, J. André, A. Simonnet, G. Maisonnier, J. Brac. Optimal design and dimensioning of Hydrogen transmission distribution pipeline networks. HEC Discussion Paper 2009 03/01, Mars 2009.
M-M. Ould Sidi, S. Hayat, S. Hammadi, P. Borne. A novel approach to developing and evaluating regulation strategies for urban transport disrupted networks. International Journal of Computer Integrated Manufacturing Vol. 21, No 4, June 2008, pp 480-493.
M-M. Ould Sidi, S. Hammadi, S. Hayat, P. Borne. Urban transport networks regulation and evaluation: A fuzzy evolutionary approach. IEEE transaction on systems mans and cybernetics SMC Part A: Systems and Humans, Vol. 38, No 2, March 2008, pp 309-318.