Use and Adequacy of Computer Paradigms to Simulate Bioinspired

Synthetic Landscape Ecologies

Jean Le Fur

, Pape Adama Mboup

and Moussa Sall

Institut de Recherche pour le Développement (IRD), Centre de Biologie pour la Gestion des Populations (CBGP),

Campus Baillarguet, CS 30016, F-34988 Montferrier-sur-Lez, France

Lab. Informatique, Faculté des Sciences et Techniques, Univ. Cheikh Anta Diop (UCAD), Dakar, Senegal

Dépt. Informatique, Univ. G. Berger/Saint-Louis Sénégal and Lab. IRD-BIOPASS, Campus Bel-Air, Dakar, Senegal

Keywords: Bio-inspiration, Ecoinformatics, Natural Computation, Synthetic Ecology, Robustness, Object-oriented

Programming, Agent-based Model, Rodent.

Abstract: In the field of ecoinformatics, synthetic ecologies approach intends to reproduce in silico the architecture and

functioning of a real domain 'as it works'. When the systems to be represented are complex, bioinspiration is

proposed as a relevant approach to develop robust modelling.

This study aims to develop, using basic object-oriented paradigms, and in collaboration with biologists, a

comprehensive synthetic ecology about a given application domain: wild rodents’ population dynamics. To

address the complexity of the field, the architecture is gradually grown and shaped from integration of

successive and diverse case studies modelling in the application domain. Developed for more than ten years,

the same model has made it possible to represent historical, cellular and/or ecological processes at the scales

of a country, a region, a city or a laboratory as well as a diversity of interacting living beings.

Results present how principles such as composition, aggregation, inheritance, generalization have been used

to elaborate a synthetic ecology. These paradigms altogether constitute a rich, and improvable, toolbox

offering a varied set of possible uses to formalize bioinspired landscape or ecologies.

1 INTRODUCTION

Bioinspiration in the more common sense is

considered as using natural solutions to solve

engineering questions. Conversely, in the field of bio-

socio-ecological modelling, one's aim is to use

computer or formal engineering to question natural

processes. The latter models often are abstraction of

the processes under investigation. For example,

differential equation sets can be used to model

epidemic process such as in the classical susceptible-

infected-recovered (SIR) model (e.g., Chen et al.,

2020). In some cases, bioinspired algorithms are also

used to formalize natural contexts. Several

approaches belonging to so-called natural computing

(Kari and Rosenberg, 2008) or ecoinformatics (Yao,

2006) are developed in that direction such as cellular

automata, genetic programming, swarm intelligence,

artificial immune systems… In this field of modelling

too, this bioinspired approach may be not related to

the biological processes studied. For example, neural

networks may be used to describe a hierarchically

structured ecosystem (Olden et al., 2006), or a

bioinspired cell model (P-system) be developed to

study a trophic web (Colomer et al., 2011) … In any

of these directions there is generally an assumed gap

between the concrete problem to solve and the model

formalism (strategic models, Holling 1966 in Evans

et al., 2013). In the case of bio-ecological modelling

this gap however may restrain each model application

to the particular use case it is fitted to or trained on

(Svoray and Benenson, 2009). This rules out an

explicative description that could be generalized in

other contexts or produce relevant forecasting even in

a changing context. Model’s robustness may there be

questioned.

On the other hand, tactical models (Holling 1966

in Evans et al., 2013) are focused on prediction and

robustness. For this purpose, their aim is to capture

the mechanisms that govern real-world dynamics

with a virtual copy working the same way (Svoray

and Benenson, 2009). Such approach can be found in

synthetic biology, a field dominated by research at the

microbiological level (Dunham, 2007, Shou et al.,

2007). At the scale of individuals, the so-called

Functional-Structural Plant Modelling or FSPM

154

Le Fur, J., Mboup, P. and Sall, M.

Use and Adequacy of Computer Paradigms to Simulate Bioinspired Synthetic Landscape Ecologies.

DOI: 10.5220/0010601101540162

In Proceedings of the 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications (SIMULTECH 2021), pages 154-162

ISBN: 978-989-758-528-9

(Buck-Sorlin, 2013) integrates biology and

physiology to reproduce plant growth (Dejong et al.,

2011). At a larger scale the emerging field of

synthetic ecology proposes bioinspired solutions in a

large range of research domains such as city planning,

sedimentation modelling (Cantrell and Holzman,

2014). In Ecology or natural landscape issues,

DeAngelis and Mooij (2003) also proposed a so-

called “mechanistically rich” approach in which a

maximum of the entangled factors producing the

natural dynamics studied could be integrated.

Developing such promising approaches supposes

a focus on robustness. Indeed, it implies accounting

for a diversity of items with highly heterogeneous

distinct behaviours and need tough computation to

manage this complexity.

On the other side, Nature appears as the one and only

robust ‘model’ where drivers are universal, where

processes and components function altogether and

resist to change. Bioinspired computation may be

then one possible direction to elaborate robust

ecological or landscape models. Since Nature is

indeed the only robust functioning system, the more

one uses the same processes for computation, the

more one can produce robust analogies and hence

simulations, of the worlds investigated.

In this study we propose to use computer

paradigms, in collaboration with bio-ecologists, to

mimic the known behaviour of an application domain

in Ecology with the objective to produce (i) the more

robust possible model and (ii) a comprehensive tool

for interaction between biologists and modelling.

Following Wirth (2006), Object-Oriented

Programming was selected as the privileged approach

to develop a model as closed as possible to what is

known of the functioning of Nature.

We first present our approach to develop a

synthetic ecology that could be as bioinspired as

possible. The structure of the resulting architecture is

then described by indicating the computational

paradigms, derived mainly from object-oriented

formalism, that have been shown as the most relevant

for mimicking the known functioning of ecosystems.

The advantages and disadvantages of object-oriented

programming from this point of view are discussed.

2 MATERIAL AND METHOD

2.1 Model Purpose

The objective of this work is the long term elaboration

of a generic model of wild rodent ecology. The aim is

primarily to provide dynamic simulations where

knowledge coming from multidisciplinary thematic

like Bio-Eco-Sociology, Geography, etc. can be

articulated. The expected outcome would be a

relevant tool with which specialists could compare

simulations with most indicators and knowledge they

are accustomed.

Two specifications found the approach:

A) The model must be as comprehensive as

possible to take into account the various sources of

fluctuations at stake in the ecosystems or landscapes

under investigation.

B) The model must hence be robust to multiple

contexts. Robustness would permit also to instantiate

a model that could be queried from multiple

multidisciplinary points of view (i.e., provide the

variety of indicators with which bio-ecologist are

used to).

To achieve (or rather go in the direction of) this

objective, we have retained as a principle that Nature

is the unique example of architecture and function

that fulfils robustness and exhaustiveness. Hence, the

greater the fidelity to the Natural 'model', the greater

the pledge of robustness. We therefore sought to

reproduce the available bio-ecological knowledge

with the most bioinspired modelling schemes

possible.

2.2 Context: Research in Rodents’

Bio-ecology and Epidemiology

This long-term project was carried out in a laboratory

of biology and ecology within a team of scientists

specialised in wild rodents, mainly in West Africa

(Granjon and Duplantier, 2009). These populations

are studied as pests (crops or dwellings) or as

reservoirs of pathogens involved in numerous

epidemics.

A preliminary investigation was first carried out

by mean of individual interviews with biologists

belonging to or associated with this laboratory. The

interviews aimed to identify the characteristics of the

knowledge domain (rodents) to be represented. The

content of five interviews was then reified and

integrated to identify the different components,

processes, variables, indicators implemented in

rodent ecology in multiple contexts.

The result of this work (Le Fur, 2014) highlighted

a rich field of knowledge with a great diversity of:

- constituents with, particularly, numerous species

interacting,

- processes and scales, especially spatial ones, to

be considered,

- approaches such as for example Agroecology,

Biogeography, community Ecology, Eco-

Use and Adequacy of Computer Paradigms to Simulate Bioinspired Synthetic Landscape Ecologies

155

Immunology, landscape or population Genetics,

Phylogenetics, Physiology,

- indicators used and the way they are obtained to

observe the ecosystems concerned.

2.3 Approach: Growing the Model as a

Heuristic

Given this complexity it was not possible to design

‘from scratch’ a simulator capable of representing

most aspects of a synthetic ecology. We therefore

chose, as a heuristic, to make grow the model by

successively formalizing concrete case studies. Each

new case study brings new facets and new constraints

to be resolved in order to account for field knowledge.

The iterative integration of various cases then makes

it possible to gradually evolve model structures,

function as well as model parameters and progress in

the development of a synthetic model. This approach

was also used as a testbed for the robustness of the

model as it was developed (see below).

Each case study was chosen in the first place

according to interest of biologists for a model of this

type in their field. Beyond that, in the modelling

project, we tried to select the most different studies

possible, however still in the field of bio-ecology of

wild rodents, so as to test the robustness of the model

to various contexts: each particular situation or use

case, once integrated, constitutes in turn a constraint

for the preceding ones and has to be accounted within

the code.

The model was developed in Java using the agent-

based Repast-Simphony platform (North et al., 2013)

and using Eclipse features for refactoring operations

in particular.

Table 1 and the text below provide a description

of the main contribution to the model for six distinct

use cases out of the 11 developed throughout the

whole project.

- A first study in a dynamic agricultural landscape

made it possible to set up the notion of nested

spaces in which similar space cells can be

grouped to compose a higher level entity (a field,

a road, etc.).

- The second case study looked at representing the

genetic process of reproduction, leading to the

production of one being from two others. The

model represents the genetic process including

gene, chromosome, …, meiosis, fusion and

crossing of chromosomes, rejection of

aberrations, transfer of genes. The model was

validated by reproducing the results of an animal

facility experiment (Comte, 2012) and was

generalized to all case studies.

- These first stages of the model made it possible

to implement and refine a study on a rodent

trapping experiment where we could carry out

‘reality-inspired’ formalization of rodent traps

scientifically scattered in the simulated landscape.

Bio-inspiration permitted here to reproduce

straightforwardly the trapping protocol and

produce the necessary ecological and genetic

Table 1: Main characteristics of the most distinctive case studies successively formalized to elaborate the model.

SIMULTECH 2021 - 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

156

outputs that were necessary to reproduce and

compare the composite estimators familiar to

environmentalists (Diakhate et al., 2014).

- The following studies were interested in the

application and the adaptation of the model to

much larger scales, it was thus possible to

represent the historical evolution of commercial

transport on the scale of a century and a whole

country and study the probability of rodents

(black rats) boarding commercial vehicles (boats,

trains, trucks) to gradually colonize the entire

country. The model has been refined here by

taking into account human agents as well as

various adaptations such as graph routes (Mboup

et al., 2015a, 2015b) to represent their moves.

- Another implementation has been developed to

account for the behaviour of rodents in a Sahelian

savannah in Africa. The hierarchy of behaviours

has been reorganized here using knowledge from

biologists. Here we have taken into account

predation (owls that eat gerbils) and circadian

activity (cycles of human and animal activity

different between night, day, dawn and dusk).

- Finally, a recent study (Sall et al., In prep.)

Looked at the spread of epidemics (zoonoses)

linked to urban mice and their proximity to

humans (commensal animals). The epidemic

chain has been formalized by integrating biting

ticks and pathogenic bacteria, cats and human

activity at high resolution. This choice has led to

an interest in particular phenomena such as the

ability of mice to walk along residential walls to

move around (Sall et al., 2019).

2.4 Use of Computer Paradigms

With each addition of a new case study, the model has

been reworked and reshaped according to new

features and constraints introduced. This work mainly

consisted in generalization that is factorizing

attributes or methods shared between several studies

and move them up to an upper parent class (owl nest

and rodent burrow generalized into animal home for

example).

One important step in this reworking has been

also the repeated use of refactoring. As we progressed

in understanding the emerging architecture, the

'natural' meaning of variables and methods was

refined to fit as much as possible to biological

knowledge. Refactoring hence gradually improved

consistency, readability, genericity and,

consequently, maintenance of the growing model and

the backward compatibility with all formalized use

cases.

Finally, each behaviour formalized in the model

was questioned, with the help of biologists, as

possibly generic. This led to enrichment of the class

hierarchy with encapsulation of natural functioning at

the specific level at which they occur.

Depending on the specific needs, encapsulation,

delegation, aggregation, polymorphism, interfacing

and other paradigms of the sort have been used to

reproduce by analogy the structures known as

occurring in Nature and elaborate a clean

architecture. In this process, apart from

generalization, three particular object-oriented

programming paradigms have been focused on to

elaborate the model: composition, aggregation and

inheritance (see results section).

At the end of this compilation of the case studies,

1,143 minor or major reconfigurations (version

commits) of the model architecture and function were

carried out. The final source code represents 22,248

lines (including data input, data retrieval, and display

output) of which 9,066 lines of code (41%) for the

business model which constitutes the synthetic

ecology itself. These 9,066 lines bring together 98

classes, 149 attributes, 181 relations and 761

methods. Regarding the latter, they contain 168

(22%) calls to super and 175 (23%) overridden

methods.

3 RESULTS

The architecture obtained at the end of the study is

presented in more detail (UML diagrams) in Le Fur

et al. (2017). In this article we focus on programming

paradigms that have been useful in advancing towards

a bioinspired synthetic ecology. To develop and

refine the code in this direction, three programming

paradigms were found particularly relevant:

composition, aggregation, inheritance.

3.1 Composition

Composition was used in this model for two generic

structures of ecosystems and landscapes.

The first is bio-inspired and straightforward to

implement: genomes are known and described as

composition and they were formalized in this way.

Thus, during mating of any two reproducing agents,

we obtain the following functional structure:

genome diades (paired chromosomes)

chromosomes genes.

Use and Adequacy of Computer Paradigms to Simulate Bioinspired Synthetic Landscape Ecologies

157

The composition principle has made it here

possible to computationally reproduce the

functioning of a genome in a manner analogous to

known reality.

The second use of composition concerns the

formalization of space into successive entities such as

"cell region landscape" or "cell cage

animal facility"... However, this choice is here

an arbitrary classification linked to observation and

which does not correspond to a genuine architecture

of Nature.

3.2 Aggregation and Recursion

The principal use of aggregation in a bio-inspired

perspective has been developed to account for a first

fundamental organization of Nature. Nature is

universally a nested system ranging from the smallest

quantum particle to the Universe. Within this system

each level arises from the emergence of interactions

between components of the lower level: atom -

molecule - cell - organ - organism - population…

These different levels are hence embedded within

each other. At the scale of an ecosystem or a

landscape each component is always simultaneously

a container for other components and itself a content.

To reflect this architecture, all agents have gradually

inherited from an abstract class (‘Container’)

implementing an interface of the same name (Figure

1).

Figure 1: The minimal interface finally obtained for any

object, included agent, sharing roles of container and

content. Methods underlined recursively run through the

cascade of containers contained (see Figure 3).

Upon completion of the case studies integration,

the Container class becomes a founding class, close

to the root of the business model, and from which all

objects and agents inherit. This thus made it possible

to 'naturally' formalize a rodent in a culture (in its

meal), a prey in the talons of an owl, a bacterium in a

tick or an embryo inside an animal (Figure 2).

Figure 2: Selected examples of embedded chains of

containers currently observable in the model’s simulations.

Each item is a Java class implementing the Container

interface.

This implementation allowed us to use recursion

to propagate global changes throughout the chain of

containers and get in this way a parsimonious and

robust management of the system (Figure 3).

Figure 3: Example (Java code) of recursion used for

containers aggregation management: propagation of a

global change to all occupants in a Container.

3.3 Inheritance

A second constitutive architecture of Nature was used

to structure the model which results from the theory

of evolution (Darwin, 1859). Living species diversify

over time by inheriting the fundamental

characteristics of their ancestors while having new

functions or structures that are specific to them.

Organisms are therefore organized according to a

single phylogeny unique to the living world and

within which each individual of any species has

genetically inherited all the specific properties of its

ancestors. The object oriented principle of inheritance

is a paradigm working like a phylogeny. Taxonomic

hierarchy (Marcos and Cavero, 2002) was thus

chosen as a robust approach to account in a sparing

way of biodiversity as it is naturally constructed. This

approach has led to the definition of a tree structure

of genomes delegated to that of the modelled agents

(see below). Each genome provides specific

characteristics to agents that bare them.

Each genome is thus analogous to a knowledge

base associated with any agent and allowing it to

achieve its growth, reproduction, life cycle ... with the

proper parameters (speed, litter size, weaning age…)

of its species. Over the course of the case studies, and

without limitation, it has been possible to characterize

plants, mammals, animals, birds, arthropods, bacteria

(Figure 4), without calling into question this

architecture. These characteristics are transmitted at

each act of reproduction occurring in a simulation,

whether it is a rodent, bird, etc.

Coupled with the preceding genome hierarchy,

and as a founding principle in agent-based modelling,

the inheritance paradigm is also the core of the

functional model within which agents’ behaviours

and interaction are coded. It was elaborated here in

/** Recursively set the value passed to self

and contained containers field */

public void setAnyValue(double a) {

this.myField = a;

for (Container oneContainer : this.getContainerList())

oneContainer.setAnyValue(a);

}

SIMULTECH 2021 - 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

158

Figure 4: The genome phylogeny model part that is arising from the sole compilation of case studies on wild rodents ecology.

Genomes objects from the different classes a delegated to the corresponding type of agent in the functional hierarchy (see

Figure 5).

conjunction with biologists to separate the

biologically functional levels and dispatch their

characteristics in accordance with current knowledge

(Figure 5). Each level was progressively

distinguished and given properties bound to bio-

ecological knowledge; the class Organism manages

genome, Animal manages deliberation and move,

Amniote manages reproduction, … This arborescence

also emerged progressively through use cases

modelling and model refactoring and restructuring.

4 DISCUSSION

4.1 Genericity, Robustness and

Reusability

Developed for ten+ years, the model has thus made it

possible to represent a laboratory experiment,

historical processes at the scale of a country,

intracellular or ecological processes at the scales of a

landscape, country or city. By trying to systematically

represent Nature as it is known, and thanks to the

process of generalization, this approach makes it

possible to reach a robust modelling allowing to

formalize and make interact in the same shared

schema a great diversity of living beings (humans,

rodents, birds, plants, parasites, predators).

The resulting architecture remains consistent and

extensible. It permits, almost straightforwardly, to put

a tree in a landscape, then a nest in the tree and an owl

in the nest. One can also address complex issues such

as the occupation of an organism by a parasite

entering and leaving its organic container. Moreover,

the scheme improves continually. As an example, the

very first use case started with a reduced set of

possible desires (forage, reproduce) used in the

deliberation scheme of rodent agents; the model then

gradually enriched with, among others, new desires

such as resting, wandering, dispersing, flee, hide,

suckle... Over time, the relevance of the simulations

for each and every specific case studies thus

continuously increases.

4.2 Paradigms Adequacy

The functional hierarchy that has been established

uses nothing more than the classical paradigms of the

object-oriented approach. But this tree structure also

revealed itself, with the collaboration between

biologists and modellers, biologically relevant. The

robustness obtained makes it possible, for example, to

Spermatophyta

Rodentia

Amniota

Acaria

Animalia

Eucaryota

GenomeLUCA

GerbillusNigeriae

HomoSapiens

RattusRattus

Borrelia

Ty t o A l b a

MusMusculus

MicrotusArvalis

Acacia

Balanites

Fabacea

Poacea

MNatalensis

MErythroleucus

Mas tomys

bush

wild grass

tick

bacteria

crop

barn owl

house mouse

black rat

prairie vole

african

rodent

desert rat

Last Universal

Common Ancestor

rodent

animal

plant

tree

raise its

eggs

pairs of

chromosomes

Use and Adequacy of Computer Paradigms to Simulate Bioinspired Synthetic Landscape Ecologies

159

Figure 5: Comprehensive functional hierarchy arising from 11 case studies: Illustrative example of classes involved in a house

mouse agent behaviour (12 classes out of 98 in the business model). The number of lines of code (right) stands here as a

proxy for the amount of behaviours and functions (left) dedicated to each level. A house mouse in this case owns the whole

set of behaviours exposed. NDS: Nearly Decomposable System (Simon, 1962).

integrate new types of organisms without calling into

question the model. It is also likely that, when

retaining only the uppermost abstract classes and

interfaces, the model could be dedicated to simulate

natural systems entirely distinct from rodent bio-

ecology.

The computational paradigms that have been

found efficient to copy natural mechanisms

(composition, recursion, aggregation, etc.) come

from a wide range of possible implementations that

could be exploited for bioinspired modelling. Each is

not used systematically but whenever necessary and

appropriate. They thus constitute a combination of 'ad

hoc' formalisms which are available as in a toolbox to

represent in a bioinspired way particular aspects of

Nature.

However, certain modes of operation of Nature

seem difficult to take into account. This is the case,

for example, of mechanisms operating

simultaneously on several spatial and temporal scales.

A study with this same model (Mboup et al., 2017)

showed that it was possible to develop algorithms

formalizing this mode of operation. But addressing

this issue leads to overly sophisticated algorithms that

move away from the bioinspired issue.

Finally, the question arises of the relevance or the

need to use multiple inheritance to develop a

synthetic ecology. For example, in this study, rodents

can be colonial, fossorial, domestic, commensal,

circadian (Major functions, Figure 5) with different

combinations depending on the species. For a bio-

inspired model, to this "logic" of Nature should

correspond an adapted computer formalism. In the

context of the Java language that was chosen at the

start of the project, the interface paradigm is not

sufficient to account for the multiple inheritance of

behaviours because it requires code duplications.

However, implementations do exist with the so-called

interface default methods of Java 8 (Mohnen, 2002)

that could be used.

Achieving a synthetic ecology using a language

dedicated to the management of multiple inheritance

(e.g., Malayeri, 2009) could thus constitute an

approach to be favoured for the development of

bioinspired simulations of ecologies or landscapes.

However, multiple inheritance gives rise to

inconsistencies such as the diamond problem (Truyen

et al., 2004). These problems may question its

relevance to mimic a real functioning of Nature

which, for its part, does not present any

inconsistencies.

5 CONCLUSION

Computer paradigms and in particular those related to

object-oriented formalism all together constitute an

SIMULTECH 2021 - 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

160

effective toolbox to advance towards the

development of bioinspired synthetic ecologies and

landscapes. They make it possible to imitate multiple

forces or principles that Nature uses to achieve a

functional and complete World. This toolbox of

paradigms and formalisms is always intended to be

improved, for example by facilitating the

dissemination and secure use of multiple inheritance.

Research on bioinspiration can then be a guide to

direct new research in this direction.

ACKNOWLEDGEMENTS

The authors would like to thank the members of the

“rodent group” of the Centre for Biology for

Population Management (CBGP) as well as the

members of the BioPASS laboratory for their

essential support in the development of this study.

REFERENCES

Buck-Sorlin G. (2013) Functional-Structural Plant

Modeling. In: Dubitzky W., Wolkenhauer O., Cho KH.,

Yokota H. (eds) Encyclopedia of Systems Biology.

Springer, New York, NY. https://doi.org/10.1007/978-

1-4419-9863-7_1479

Cantrell, B., & Holzman, J. (2014). Synthetic Ecologies:

protocols, simulation, and manipulation for

indeterminate landscapes.

Chen, Y. C., Lu, P. E., Chang, C. S., & Liu, T. H. (2020).

A time-dependent SIR model for COVID-19 with

undetectable infected persons. IEEE Transactions on

Network Science and Engineering, 7(4), 3279-3294.

Colomer, M. À., Margalida, A., Sanuy, D., Pérez-Jiménez,

M. J. (2011). A bio-inspired computing model as a new

tool for modeling ecosystems: the avian scavengers as

a case study. Ecological modelling, 222(1), 33-47.

Comte A. (2012) Caractérisation des barrières à

l'hybridation de deux espèces jumelles de rongeurs

africains du genre Mastomys. Etude par simulation

multi-agents à partir de deux expériences in situ. Rapp.

M2 Ecologie-Biodiversité Spécialité Biodiversité

Evolution Parcours Génétique et Biodiversité, june

2012, 44p.

Darwin C (1859) On the origin of species by means of

natural selection, or the preservation of favoured races

in the struggle for life. John Murray, London.

DeAngelis, D.L., Mooij, W.M., 2003. In praise of

mechanistically rich models. In: Canham, C.D., Cole,

J.J., Lauenroth, W.K. (Eds.), Models in Ecosystem

Science. Princeton University Press, Princeton, New

Jersey, pp. 63–82.

Dejong, T., Da Silva, D., Vos, J., and Escobar-Gutiérrez, A.

(2011). Using functional–structural plant models to

study, understand and integrate plant development and

ecophysiology. Annals of Botany 108, 987-989.

Diakhate, E.H.M., Diouf, N., Granjon, L., Konate, K.,

Mboup, P.A. et J. Le Fur (2014) Modélisation et

simulation multi-agents d'un protocole de capture-

marquage-recapture d'une population de rongeurs

sauvages dans la réserve de Bandia (Sénégal). 12th

African Conference on Research in Computer Science

and Applied Mathematics (CARI), Saint-Louis Sénégal,

20 au 23 octobre 2014: 43-54.

Dunham, M. J. (2007). Synthetic ecology: a model system

for cooperation. Proceedings of the National Academy

of Sciences, 104(6), 1741-1742.

Evans, M. R., Grimm, V., Johst, K., Knuuttila, T., De

Langhe, R., Lessells, C. M., ... Benton, T. G. (2013). Do

simple models lead to generality in ecology?. Trends in

ecology & evolution, 28(10), 578-583.

Granjon, L., Duplantier, J. M. (2009) Les rongeurs de

l'Afrique sahélo-soudanienne. IRD Editions,

Publications Scientifiques du Muséum - Collection

Faune et Flore tropicales 43. ISBN IRD: 978-2-7099-

1675-2

Holling, C.S. (1966) The strategy of building models of

complex ecological systems. In Systems Analysis in

Ecology (Watt, K.E.F., ed.), pp. 195–214, Academic

Press.

Kari, L., Rozenberg, G. (2008). The many facets of natural

computing. Communications of the ACM

, 51(10), 72-

83.

Le Fur, J. (2014) Du foisonnement des disciplines à la

recomposition d'une réalité partagée Elaboration

d'une structure de modélisation dédiée à l'intégration

de connaissances disciplinaires. XXIeme journées

de Rochebrune "Multi-trans-interdisciplinarité".

Rencontres interdisciplinaires sur les systèmes

complexes naturels et artificiels. 19-25 janvier 2014,

(https://hal.ird.fr/ird-03182988).

Le Fur, J., Mboup, P.A. and M. Sall (2017) A Simulation

Model for Integrating Multidisciplinary Knowledge in

Natural Sciences. Heuristic and Application to Wild

Rodent Studies. 7th Internat. Conf. Simul. and Model.

Method., Technol. and Applic. (Simultech), Madrid,

july 2017

Malayeri, D., Aldrich, J. (2009). CZ: multiple inheritance

without diamonds. ACM SIGPLAN Notices, 44(10), 21-

40.

Marcos, E. and Cavero, J.M. (2002) Hierarchies in Object

Oriented Conceptual Modeling. In Bruel, J. and

Bellahsène Z. (Eds.): OOIS 2002 Workshops, LNCS

2426, pp. 24-33, 2002.

Mboup, P.A., Konaté, K., et Le Fur, J. (2017) A multi-

world agent-based model working at several spatial and

temporal scales for simulating complex geographic

systems. Internat. Conf. Comput. Sci. (ICCS), Zurich,

12-14 juin 2017. Procedia Computer Science 108C:

968–977 doi:10.1016/j.procs.2017.05.090

Mboup, P.A., Konaté, K., Handschumacher, P. et J. Le Fur

(2015). Des Connaissances au Modèle Multi-agents

par l’approche orientée événement : construction d’un

simulateur de la colonisation du rat noir au Sénégal par

Use and Adequacy of Computer Paradigms to Simulate Bioinspired Synthetic Landscape Ecologies

161

les transports humains sur un siècle. In actes Colloque

National sur la Recherche en Informatique et ses

Applications, Thiès, Sénégal, 8p.

Mboup, P.A., Mboup, M.L., Konaté, K., Handschumacher,

P., et Le Fur, J. (2015). Optimisation de l’utilisation de

l’algorithme de Dijkstra pour un simulateur multi-

agents spatialisé. In: World Congress on Information

Technology and Computer Applications (WCITCA),

Hammamet, Tunisia, pp. 1-6. doi:10.1109/WCITCA.20

15.7367061

Mohnen, M. (2002) Interfaces with default implementa-

tions in Java. Proc. workshop on Intermediate

representation engineering for virtual machines, June

2002 Pages 35–40

North, M.J., Collier, N.T., Ozik, J. Tatara, E., Altaweel, M.,

Macal, C.M., Bragen, M. and P Sydelko (2013)

Complex Adaptive Systems Modeling with Repast

Simphony. Complex Adaptive Systems Modeling,

Springer, Heidelberg, FRG (2013). https://doi.org/

10.1186/2194-3206-1-3.

Olden, J. D., Poff, N. L., Bledsoe, B. P. (2006).

Incorporating ecological knowledge into

ecoinformatics: An example of modeling hierarchically

structured aquatic communities with neural networks.

Ecological Informatics, 1(1), 33-42.

Sall, M., Dembélé, J.M. and J. Le Fur (2019) An hybrid

algorithm to simulate mice following residential wall.

Proc. 8th Internat. Conf. Simul. and Model. Method.,

Technol. and Applic. (Simultech), ScitePress Publ.:

368-375

Shou, W., Ram, S., Vilar, J. M. (2007). Synthetic

cooperation in engineered yeast populations.

Proceedings of the National Academy of Sciences,

104(6), 1877-1882.

Simon, H. A., 1962. The Architecture of Complexity. Proc.

American Philosophical Society, 106(6):467-482.

Svoray, T., Benenson, I. (2009). Scale and adequacy of

environmental microsimulation. Ecological

Complexity, 6(2), 77-79.

Truyen, E., Joosen, W., Nørregaard Jørgensen, B. and

Pierre Verbaeten (2004) A Generalization and Solution

to the Common Ancestor Dilemma Problem in

Delegation-Based Object Systems. Proceedings of the

2004 Dynamic Aspects Workshop, nos 103–119, 2004

Wirth, N. (2006) Good Ideas through the Looking Glass.

Computer, vol. 39, no. 1, pp. 28-39, 2006.

Yao, X., Liu, Y., Li, J., He, J., Frayn, C. (2006). Current

developments and future directions of bio-inspired

computation and implications for ecoinformatics.

Ecological informatics, 1(1), 9-22.

SIMULTECH 2021 - 11th International Conference on Simulation and Modeling Methodologies, Technologies and Applications

162