first commit

1_Buildings and roads environment.gaml

+/***
+* Part of the SWITCH Project
+* Author: Nathan Coisne, code inspired from Patrick Taillandier
+* Tags: gis, OSM data
+***/
+
+model switch_utilities_gis
+/*
+ * Genere les fichiers buildings.shp et roads.shp pour la ville de Marseille
+ * 
+ * Données requises pour générer les fichiers buildings.shp et roads.shp d'une ville quelconque : 
+ * https://download.geofabrik.de/europe/france.html pour le shp 'initial' des builings
+ * https://geoservices.ign.fr/documentation/diffusion/telechargement-donnees-libres.html#bd-topo pour le shapefile à partir duquel on extrait les attributs qu'on donne aux buildings initiaux
+ * Les frontières de la zone considérée
+ * 
+ */
+ 
+global {
+	
+	string dataset_path <- "../includes/Marseille/";
+	
+	//define the bounds of the studied area
+	file data_file <-shape_file(dataset_path + "boundary_Marseille.shp");
+	
+	//define the initial building shapefile, extracted from OSM database 
+	file shape_file_buildings <- shape_file(dataset_path + "buildings_marseille.shp");
+	
+	//define the OSM points of interest buildings, for more types
+	file shape_file_pois <- shape_file("../includes/Marseille/points_of_interest.shp");
+	
+	//define the building shapefile with many types, extracted from IGN database
+	file ign_file <-  shape_file(dataset_path + "buildings_ign.shp");
+	
+	//define the initial building shapefile, extracted from OSM database 
+	file shape_file_roads <- shape_file(dataset_path + "roads_marseille.shp");
+		
+	
+    list<string> shop_places <- ["commercial", "kiosk", "chapel", "church", "service", "Commercial et services", "Religieux", "religious"];
+    list<string> sport_places <- ["Sportif"];
+	
+	
+	float min_area_buildings <- 20.0;
+	int nb_for_building_shapefile_split <- 50000;
+	int nb_for_road_shapefile_split <- 20000;
+	
+	
+	list<string> living_places <- ["house", "apartments", "dormitory", "hotel", "residential", "Résidentiel"];
+    list<string> work_places <- ["industrial", "office", "construction", "garages", "hospital"];
+    list<string> study_places <- ["university", "college", "school"];
+    list<string> leisure_places <- ["commercial", "kiosk", "chapel", "church", "service", "Sportif", "Commercial et services", "Religieux", "religious"];
+	
+	bool parallel <- true;
+	
+	bool buildings <- true;
+	bool roads <- false;
+	
+	
+	geometry shape <- envelope(data_file);
+	
+	graph the_graph;
+        
+    map<road,float> road_weights;
+	
+	init {
+		write "Start the pre-processing process";
+		
+		list<int> list_of_id;
+		if buildings{
+			create Building from: shape_file_buildings with: [type::get ("type"), id::int(get("osm_id"))]{
+				if not (self overlaps world) {
+					do die;
+				}
+				list_of_id << id;
+				if (type != nil and type != ""){
+					types << type;
+				}
+	   		}
+			
+			write "Number of buildings created : "+ length(Building);
+			
+			ask Building where (each.shape.area < min_area_buildings) {
+				do die;
+			}
+			
+			write "Small building removed";
+			
+			
+			create Building_ign from: ign_file {
+				if not (self overlaps world) {
+					do die;
+				}
+			}
+			
+			write "Number of buildings ign created : "+ length(Building_ign);
+			
+			ask Building parallel: parallel {
+				list<Building_ign> neigh <- Building_ign overlapping self;
+				if not empty(neigh) {
+					Building_ign bestCand;
+					if (length(neigh) = 1) {
+						bestCand <- first(neigh);
+					} else {
+						bestCand <- neigh with_max_of (each inter self).area;
+					 	if (bestCand = nil) {
+					 		bestCand <- neigh with_min_of (each.location distance_to location);
+					 	}
+					 	if (bestCand = nil){
+					 		write("aucun building correspondant a ce building ign");
+					 	}
+					 }
+					 if (bestCand.USAGE1 != nil and bestCand.USAGE1 != ""){ 
+					 	types << bestCand.USAGE1;
+					 }
+					 else if (bestCand.USAGE2 != nil and bestCand.USAGE2 != ""){
+					 	types << bestCand.USAGE2;
+					 }
+
+					 if (bestCand.HAUTEUR != nil and bestCand.HAUTEUR > 0){ 
+					 	height <- bestCand.HAUTEUR;
+					}
+				}
+			}
+			
+			create Building from: shape_file_pois with: [id::int(read("osm_id")), type::read("fclass")]{
+    		if not (self overlaps world){
+    			do die;
+    		}
+    		
+    		if not (id in list_of_id){
+    			
+    			list_of_id << id;
+    			
+    			types_str <- type;
+	    		types << type;
+	    		
+    		}
+    		
+    		else{ // fclass devient le type du bâtiment déjà existant
+    			Building real_building <- Building  first_with (each.id = self.id); 
+    			real_building.type <- type;
+    			real_building.types << type;
+    			real_building.types_str <- type + "," + real_building.types_str;
+    			do die;
+    		}  		
+    	}
+				
+				
+			ask Building where empty(each.types){
+				do die;
+			}
+				
+			ask Building parallel: parallel {
+				type <- first(types); //si le building a déjà un type dans le shapefile initial : on le garde en position 1. Sinon on prend celui de building_ign (USAGE1 ou USAGE2)
+				
+				types_str <- type;
+				
+				if (length(types) > 1) {
+					loop i from: 1 to: length(types) - 1 {
+						types_str <-types_str + "," + types[i] ;
+					}
+				}
+			}
+			
+			ask Building_ign{
+				do die;
+			}
+			
+			write("All buildings created and typed. Start cleaning roads");
+		}
+		
+		
+		if roads{
+			write("cleaning roads");
+			list<geometry> clean_lines <- clean_network(shape_file_roads.contents,3.0 , true, true);
+			write("road shapefile cleaned");    	
+	    	
+	    	create road from: clean_lines with: [id::int(get("osm_id")), fclass::get("fclass"), maxspeed::int(get("maxspeed"))]{
+	    		if not (self overlaps world) {
+					do die;
+				}
+    		}
+    	write "Number of roads created : "+ length(road);
+    	
+		}
+		    	
+    	if buildings{	
+			write("Saving buildings");  
+			save Building to: dataset_path + "buildings.shp" type: shp attributes: ["id"::id,"type"::type, "types_str"::types_str ,"height"::height];
+
+
+		}
+		
+		if roads{
+			write("Saving roads");  
+			save road to: dataset_path + "roads.shp" type: shp attributes: ["id"::id,"fclass"::fclass, "maxspeed"::maxspeed];
+
+		}
+	}
+}
+
+species Building_ign {
+	/*nature du bati; valeurs possibles: 
+	* Indifférenciée | Arc de triomphe | Arène ou théâtre antique | Industriel, agricole ou commercial |
+Chapelle | Château | Eglise | Fort, blockhaus, casemate | Monument | Serre | Silo | Tour, donjon | Tribune | Moulin à vent
+	*/
+	string NATURE;
+	
+	/*
+	 * Usage du bati; valeurs possibles:  Agricole | Annexe | Commercial et services | Industriel | Religieux | Sportif | Résidentiel |
+Indifférencié
+	 */
+	string USAGE1; //usage principale
+	string USAGE2; //usage secondaire
+	int NB_LOGTS; //nombre de logements;
+	int NB_ETAGES;// nombre d'étages
+	float HAUTEUR; 
+}
+species Building {
+	string type;
+	list<string> types;
+	string types_str;
+	float height;
+	int id;
+}
+
+species road  {
+	int id;
+	int maxspeed;
+	string fclass;
+		
+    aspect base {
+    draw shape color: color width: 2;
+    }
+}
+
+
+experiment generateGISdata type: gui {
+}

2_Adapt roads to cyclists.gaml

+/***
+* Part of the SWITCH Project
+* Author: Patrick Taillandier
+* Tags: gis, OSM data
+***/
+
+model switch_utilities_gis
+/*
+ * Ajoute un coefficient de vitesse correspondant à un poids dans le graphe des routes (coefficient de vitesse élevé = vitesse élevée à vélo)
+ * Ajoute la vitesse du cycliste sur chaque portion de route 
+ * Supprime les routes inaccessibles à vélo
+ */
+ 
+global {
+	
+	string dataset_path <- "../includes/";
+	
+	//define the bounds of the studied area
+	file data_file <-shape_file(dataset_path + "Marseille/boudary_Marseille.shp");
+
+	file shape_file_roads <- shape_file(dataset_path + "Marseille/roads.shp");	
+				
+	geometry shape <- envelope(data_file);
+	
+	graph the_graph;
+        
+    map<road,float> road_weights;
+	
+	init {
+		write "Start the pre-processing process";
+				
+    	
+    	create road from: shape_file_roads with: [id::int(get("id")), fclass::get("fclass"), maxspeed::int(get("maxspeed"))]{
+    		
+    		if fclass = "motorway" or fclass = "motorway_link"{do die;} //routes innacessibles aux vélos
+			
+			if fclass="trunk" or fclass="primary" or fclass="secondary" or fclass="tertiary" or fclass="bridleway" or fclass="cycleway"
+			   or fclass="trunk_link" or fclass="primary_link" or fclass="secondary_link" {
+    			speed_coeff <- 1.7;
+    			speed <- 17 #km / #h;
+    		}
+			
+			if fclass="unclassified" or fclass="residential" or fclass="service" or fclass="track" {
+    			speed_coeff <- 1.2;
+    			speed <- 12 #km / #h;
+    		}
+    		
+    		if fclass="living_street" or fclass="pedestrian" or fclass="footway" or fclass="path" or fclass="steps" or fclass="unknown" {
+    			speed_coeff <- 0.4;
+    			speed <- 4 #km / #h;
+    		}
+    		
+    	}
+    	
+		write "Number of roads created : "+ length(road);    	
+				
+		write("Saving roads");    	
+		
+		save road to: "../includes/Marseille/roads_cyclists.shp" type: shp attributes: ["id"::id,"fclass"::fclass, "s_coeff"::speed_coeff, "mean_speed"::speed, "maxspeed"::maxspeed];
+		
+		write("over");
+		
+	}
+}
+
+
+species road  {
+	int id;
+	int maxspeed;
+	string fclass;
+	
+	float speed_coeff <- 0.33;
+	float speed <- 5 #km / #h;
+		
+    aspect base {
+    draw shape width: 2;
+    }
+}
+
+experiment generateGISdata type: gui {
+	output {
+		display map type: opengl draw_env: false{
+			species road;
+		}
+	}
+}

3_Synthetic population.gaml

+/**
+* Name: Deplacementdevelos
+* Based on the internal empty template. 
+* Author: nathancoisne
+* Tags: 
+*/
+
+
+model Deplacementdevelos
+
+/* Génération d'une population synthétique de cyclistes
+ * Trois classes sociales sont représentées : travailleurs, étudiants et trajets de loisi
+ * Sources : Mobiliscope, données 'occupation des résidents par quartier'
+ * 			 INSEE household surveys
+ * 
+ * 
+ */
+
+global {
+	string city <- "Marseille";
+	bool test_population <- false; // permet de générer une population test 20 fois moins peuplée que la population réelle
+	
+	
+	file bound <- shape_file("../includes/" + city + "/boundary_" + city + ".shp");
+		
+    file shape_file_buildings; // building contenant les data osm, ign, pois
+            
+    file shape_file_quartiers; // shapefile des quartiers de la ville considérée, importée de Mobiliscope
+        
+    file residents_csv_file; // fichier contenant le nombre de résidents pour chaque quartier, importé de Mobiliscope 
+    
+    geometry shape <- envelope(bound);
+    
+    // Etalonnage d'un modèle gravitaire pour fixer les trajets des agents (leur destination): + le coeff est grand + les trajets sont courts
+    float worker_adaptative_coeff <- 0.00083;  // distance moyenne : 2000 m 
+    float student_adaptative_coeff <- 0.00095; // distance moyenne : 1600 m
+    float leisure_adaptative_coeff <- 0.00083; // distance moyenne : 2000 m
+    float delivery_adaptative_coeff <- 0.0005; // distance moyenne : 3000 m
+    
+	
+	int hour_early_leisure_start <- 11;
+	int hour_late_leisure_start <- 15;
+	
+	int hour_early_leisure_end <- 14;
+	int hour_late_leisure_end <- 18;
+	
+	int hour_early_delivery_midi_start <- 11;
+	int hour_late_delivery_midi_start <- 13;
+	
+	int hour_early_delivery_soir_start <- 18;
+	int hour_late_delivery_soir_start <- 20;
+	                            
+	list<string> living_places <- ["house", "apartments", "dormitory", "hotel", "residential", "Résidentiel"];
+		
+    list<string> work_places <- ["industrial", "office", "construction", "garages", "hospital", "service", "Commercial et services", "commercial", 
+    							 "bakery", "bank", "bicycle_shop", "bookshop", "butcher", "cafe", "car_dealership", "community_centre", "courthouse", 
+    							 "dentist", "department_store", "doctors", "fire_station", "florist", "furniture_shop", "greengrocer", "guesthouse", 
+    							 "hairdresser", "kindergarten", "kiosk", "library", "mall", "optician", "pharmacy", "police", "post_office", "prison",
+    							 "public_building", "retail", "supermarket", "town_hall", "travel_agent", "warehouse", "wastewater_plant", "water_works"];
+    							 
+    list<string> study_places <- ["university", "college", "school", "library"];
+    
+    list<string> eating_places <- ["fast_food", "restaurant"];
+    
+    list<string> leisure_places <- ["commercial", "kiosk", "chapel", "church", "service", "Sportif", "Commercial et services", "Religieux", "religious",
+    							    "bar", "buddhist", "cafe", "car_wash", "christian", "christian_catholic", "christian_evangelical", "christian_orthodox", "christian_protestant",
+    							    "cinema", "clothes", "convenience", "doityourself", "dog_park", "florist", "fort", "fountain", "garden_centre", "gift_shop", "golf_course", "jewish",
+    							    "mall", "museum", "muslim_sunni", "newsagent", "nightclub", "park", "picnic_site", "pitch", "playground", "pub", "sports_centre", "stadium",
+    							    "supermarket", "swimming_pool", "theatre"  ];
+
+    list<string> tourism_places <- ["archaeological", "arts_centre", "artwork", "attraction", "castle", "chapel", "church", "Religieux", "religious", "buddhist",
+    							    "christian", "christian_catholic", "christian_evangelical", "christian_orthodox", "christian_protestant", "fort", "fountain", "garden_centre", 
+								    "gift_shop", "graveyard", "jewish", "memorial", "monument", "museum", "muslim_sunni", "park", "ruins", "stadium", "theatre", "tourist_info", "tower"];
+   
+   
+    init{    	
+    	
+    	if city = "Marseille"{
+    		shape_file_buildings <- shape_file("../includes/Marseille/buildings.shp"); // building contenant les data osm, ign, pois
+            		            
+		    shape_file_quartiers <- shape_file("../includes/Marseille/MARSEILLE_secteurs.shp");
+		        
+		    residents_csv_file <- csv_file("../includes/Marseille//occ_nb_marseille.csv",","); //infos de population résidente dans chaque quartier
+		    
+		    matrix data <- matrix(residents_csv_file);
+	    	create building from: shape_file_buildings with: [id::int(read("id"))] {
+	
+	    		types <- types_str split_with ",";
+	    	}
+    	
+		    create quartier from: shape_file_quartiers with: [ID::int(read("CODE_SEC")), name::read("LIB"), centre::int(read("centre"))]{
+	    		if not (self overlaps world) {
+					do die;
+				}
+				
+				active_pop <- int(data[2, ID - 1]); // le nombre de travailleurs logeant dans le quartier = le nombre d'actifs présents dans le quartier à 4am
+				students_pop <- int(data[3, ID - 1]);
+				retired_pop <- int(data[5, ID - 1]);
+				inactive_pop <- int(data[6, ID - 1]);
+				without_job_pop <- int(data[4, ID - 1]);
+				
+				float coeff_surface <- (world inter self).area / self.shape.area;
+				write("id : " + ID + " coeff : " + coeff_surface);
+				
+				if coeff_surface < 0.98{
+					active_pop <- int(active_pop * coeff_surface);
+					students_pop <- int(students_pop * coeff_surface);
+					retired_pop <- int(retired_pop * coeff_surface);
+					inactive_pop <- int(inactive_pop * coeff_surface);
+					without_job_pop <- int(without_job_pop * coeff_surface);	
+				}
+				
+				if (centre = 1){
+					centre_ville <- true;
+					color <- #orange;
+				}
+				else{color <- #yellow;}
+	    	}
+    	}
+    	
+    	if city = "Toulouse"{
+    		shape_file_buildings <- shape_file("../includes/Toulouse/buildings_with_pois.shp"); // building contenant les data osm, ign, pois
+            		            
+		    shape_file_quartiers <- shape_file("../includes/Toulouse/toulouse_secteurs.shp");
+		     
+		    residents_csv_file <- csv_file("../includes/Toulouse/occ_nb_toulouse.csv",",");
+		    		    		    
+		    matrix data <- matrix(residents_csv_file);
+	    	create building from: shape_file_buildings with: [id::int(read("id"))] {
+	
+	    		types <- types_str split_with ",";
+	    	}
+		    
+		    create quartier from: shape_file_quartiers with: [ID::int(read("CODE_SEC")), name::read("LIB"), centre::int(read("centre"))]{
+		    	if not (self overlaps world) {
+					do die;
+				}
+				
+				active_pop <- int(data[2, ID - 1]); // le nombre de travailleurs logeant dans le quartier = le nombre d'actifs présents dans le quartier à 4am
+				students_pop <- int(data[3, ID - 1]);
+				retired_pop <- int(data[5, ID - 1]);
+				inactive_pop <- int(data[6, ID - 1]);
+				without_job_pop <- int(data[4, ID - 1]);
+				
+				float coeff_surface <- (world inter self).area / self.shape.area;
+				write("id : " + ID + " area : " + self.shape.area);
+				write("id : " + ID + " coeff : " + coeff_surface);
+				
+				if (centre = 1){
+					centre_ville <- true;
+					color <- #red;
+				}
+				else{color <- #blue;}
+	    	}
+    	}
+    	
+    	list<building> residential_buildings <- building where (not empty(living_places inter each.types));
+    	list<building> university_buildings <- building where (not empty(study_places inter each.types));
+    	list<building> work_buildings <- building where (not empty(work_places inter each.types));
+    	list<building> leisure_buildings <- building where (not empty(leisure_places inter each.types));
+    	list<building> eat_buildings <- building where (not empty(eating_places inter each.types));
+    	list<building> tourism_buildings <- building where (not empty(tourism_places inter each.types));
+    	
+    	// placement des agents par quartier
+    	
+    	int normalized_delivery_number;
+    	int number_of_delivery;
+    	
+    	if city = "Marseille"{number_of_delivery <- 850;}
+    	if city = "Toulouse"{number_of_delivery <- 570;}
+    	
+    	string file_path <- "../results/" + city + "/synthetic_population";
+    	if test_population{file_path <- file_path + "_test";}
+    	
+    	//
+    	loop quart over: quartier {
+    		
+    		write("population quartier " + quart.ID);
+    		float cyclist_proportion;
+    		
+    		// définition de la proportion de cyclistes parmi la population totale
+    		if quart.centre_ville{ // proportion de cyclistes en centre-ville deux fois plus élevée (3.9%) qu'en périphérie (2.2%)
+    			cyclist_proportion <- 0.039;
+    			normalized_delivery_number <- 2 * int(number_of_delivery / (length(quartier) + length(quartier where each.centre_ville)));
+    		}
+    		
+    		else{
+    			cyclist_proportion <- 0.022;
+    			normalized_delivery_number <- int(number_of_delivery / (length(quartier) + length(quartier where each.centre_ville)));
+    		}
+    		
+    		int number_of_workers;
+    		if city = "Marseille"{number_of_workers <- int(quart.active_pop * 0.013);} // source particulière : INSEE
+    		if city = "Toulouse"{number_of_workers <- int(quart.active_pop * cyclist_proportion);}
+    		
+    		int number_of_students <- int(quart.students_pop * cyclist_proportion);
+    		
+    		int number_of_leisures <- int((quart.retired_pop + quart.without_job_pop) * cyclist_proportion);
+    		 
+    		if test_population {
+    			number_of_workers <- int(number_of_workers / 20);
+    			number_of_students <- int(number_of_students / 20);
+    			number_of_leisures <- int(number_of_leisures / 20);
+    			normalized_delivery_number <- int(normalized_delivery_number / 20);
+    		}		
+    		 
+    		create worker number: number_of_workers{
+    			
+	    		living_place <- one_of (residential_buildings where (each overlaps quart));
+	    		location <- any_location_in (living_place);
+	    		
+	    		destination_place <- rnd_choice(gravitational_model(living_place, work_buildings where (each.location distance_to living_place < 15000 #m), worker_adaptative_coeff));
+	    		
+	    		int_go <- int(gauss(120, 50)); // distribution gaussienne de l'heure de départ entre 6am et 10am
+	    		go_out_hour <- int(int_go / 60) + 6;
+	    		go_out_min <- int((int_go - (go_out_hour - 6) * 60) / 5) * 5;
+	    		
+	    		int_home <- int(gauss(120, 50));
+	    		go_home_hour <- int(int_home / 60) + 15;
+	    		go_home_min <- int((int_home - (go_home_hour - 15) * 60) / 5) * 5;
+
+	        	save [go_out_hour, go_out_min, go_home_hour, go_home_min, living_place.id, destination_place.id] to: file_path + "/worker.csv" type:"csv" rewrite: false;
+
+	    	}
+	    	
+	    	create student number: number_of_students {
+				living_place <- one_of (residential_buildings where (each overlaps quart));
+				location <- any_location_in (living_place);
+	    		
+	    		destination_place <- rnd_choice(gravitational_model(living_place, university_buildings where (each.location distance_to living_place < 15000 #m), student_adaptative_coeff));
+	    		
+	    		int_go <- int(gauss(90, 40));
+	    		go_out_hour <- int(int_go / 60) + 6;
+	    		go_out_min <- int((int_go - (go_out_hour - 6) * 60) / 5) * 5;
+	    		
+	    		int_home <- int(gauss(90, 40));
+	    		go_home_hour <- int(int_home / 60) + 15;
+	    		go_home_min <- int((int_home - (go_home_hour - 15) * 60) / 5) * 5;
+	    		
+	        	save [go_out_hour, go_out_min, go_home_hour, go_home_min, living_place.id, destination_place.id] to: file_path + "/student.csv" type:"csv" rewrite: false;
+	    	}
+	    	
+	    	create leisure number: number_of_leisures {
+	    		living_place <- one_of (residential_buildings where (each overlaps quart));
+	    		location <- any_location_in (living_place);
+	    		
+	    		destination_place <- rnd_choice(gravitational_model(living_place, leisure_buildings where (each.location distance_to living_place < 15000 #m), leisure_adaptative_coeff));
+	    		
+				go_out_hour <- rnd(9, 12);
+				go_out_min <- rnd(0, 59, 5);
+				
+	        	go_home_hour <- rnd(go_out_hour + 2, 21); 
+	        	go_home_min <- rnd(0, 59, 5);
+	        	
+	        	save [go_out_hour, go_out_min, go_home_hour, go_home_min, living_place.id, destination_place.id] to: file_path + "/leisure.csv" type:"csv" rewrite: false; 
+	    	}
+	    	
+	    	create delivery number: normalized_delivery_number {
+	    		living_place <- one_of (residential_buildings where (each overlaps quart));
+		    	location <- any_location_in (living_place);
+		    	
+	    		if(rnd(1.0) <= 0.40){
+					midi <- true;
+					
+					go_out_hour_midi <- rnd(hour_early_delivery_midi_start, hour_late_delivery_midi_start - 1);
+					go_out_min_midi <- rnd(0, 59, 5);
+					
+		    		restaurant_1 <- rnd_choice(gravitational_model(living_place, eat_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		residential_place_1 <- rnd_choice(gravitational_model(restaurant_1, residential_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		
+		    		restaurant_2 <- rnd_choice(gravitational_model(residential_place_1, eat_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		residential_place_2 <- rnd_choice(gravitational_model(restaurant_2, residential_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		
+		    		restaurant_3 <- rnd_choice(gravitational_model(residential_place_2, eat_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		residential_place_3 <- rnd_choice(gravitational_model(restaurant_3, residential_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		
+		    		save [go_out_hour_midi, go_out_min_midi, living_place.id, restaurant_1.id, residential_place_1.id, restaurant_2.id, residential_place_2.id, restaurant_3.id, 
+		        	      residential_place_3.id]
+		        	  to: file_path + "/delivery_midi.csv" type:"csv" rewrite: false;
+		    		
+				}
+				
+				else {
+					midi <- false;
+					
+					go_out_hour_soir <- rnd(hour_early_delivery_soir_start, hour_late_delivery_soir_start - 1);
+					go_out_min_soir <- rnd(0, 59, 5);
+					
+		    		restaurant_1 <- rnd_choice(gravitational_model(living_place, eat_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		residential_place_1 <- rnd_choice(gravitational_model(restaurant_1, residential_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		
+		    		restaurant_2 <- rnd_choice(gravitational_model(residential_place_1, eat_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		residential_place_2 <- rnd_choice(gravitational_model(restaurant_2, residential_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		
+		    		restaurant_3 <- rnd_choice(gravitational_model(residential_place_2, eat_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		residential_place_3 <- rnd_choice(gravitational_model(restaurant_3, residential_buildings where (each.location distance_to living_place < 5000 #m), delivery_adaptative_coeff));
+		    		
+		    		save [go_out_hour_soir, go_out_min_soir, living_place.id, restaurant_1.id, residential_place_1.id, restaurant_2.id, residential_place_2.id, restaurant_3.id, 
+		        	      residential_place_3.id]
+		        	  to: file_path + "/delivery_soir.csv" type:"csv" rewrite: false;
+		    		
+				}
+	    	}
+    	}
+
+    	write("Population de cyclistes à " + city + " :");
+    	write("nombre de travailleurs : " + length(worker));
+    	write("nombre d'étudiants : " + length(student));
+    	write("nombre d'agents loisir : " + length(leisure));
+    	write("nombre d'agents livreurs : " + length(delivery));	 
+    }        
+}
+
+species building {
+	int id;
+    string type; 
+    string types_str;
+    list<string> types;
+    float height;
+    rgb color <- #gray;
+        
+    aspect base {
+    draw shape color: color ;
+    }
+}
+
+species quartier {
+	int ID;
+	string name;
+	
+	int active_pop;
+	int students_pop;
+	int retired_pop;
+	int inactive_pop;
+	int without_job_pop;
+		
+	int centre;
+	bool centre_ville; // si true : proportion de vélos 2x plus importante
+	
+	rgb color;
+	aspect base {
+    draw shape width: 6;
+    }
+}
+
+species people skills: [moving] {
+	rgb color;
+	building living_place <- nil;
+	building destination_place <- nil;
+	
+	int go_out_hour;
+	int go_out_min;
+	int go_home_hour;
+	int go_home_min;			
+	
+        
+    map<building, float> gravitational_model(building starting_place, list<building> destination_to_choose, float adaptative_coeff){
+    	// À partir d'une location de départ starting_place, choisit une destination parmi la liste destination_to_choose, selon un modèle gravitaire étalonné par adaptative_coeff
+    	 
+    	list<float> proba_list;
+    	loop build over:destination_to_choose{
+    		proba_list << exp(- adaptative_coeff * (starting_place distance_to build));
+    	}
+    		
+    	return create_map(destination_to_choose, proba_list); // retourne une map avec en clé la liste de buildings candidats, et en valeur la probabilité d'affectation du building
+    }
+    
+    
+	aspect base{
+		draw circle(6) color: color border: #black;
+	}
+}
+
+species worker parent: people {
+	rgb color <- #black;
+	
+	int int_go min: 0 max: 240;
+	int int_home min: 0 max: 240;
+}
+
+species student parent: people{
+	rgb color <- #lightgreen;
+	
+	int int_go min: 0 max: 180;
+	int int_home min: 0 max: 180;
+}
+
+species leisure parent: people{
+	rgb color <- #pink;
+}
+
+species delivery parent: people{
+	building restaurant_1;
+	building restaurant_2;
+	building restaurant_3;
+
+	building residential_place_1;
+	building residential_place_2;
+	building residential_place_3;
+	
+	bool midi;
+	
+	int go_out_hour_midi;
+	int go_out_min_midi;
+	
+	int go_out_hour_soir;
+	int go_out_min_soir;
+	
+	rgb color <- #red;
+}
+
+experiment population_generation type: gui {    
+    parameter "Earliest hour to start leisure" var: hour_early_leisure_start category: "Leisure" min: 8 max: 13;
+    parameter "Latest hour to start leisure" var: hour_late_leisure_start category: "Leisure" min: 12 max: 18;
+    parameter "Earliest hour to end leisure" var: hour_early_leisure_end category: "Leisure" min: 11 max: 17;
+    parameter "Latest hour to end leisure" var: hour_late_leisure_end category: "Leisure" min: 15 max: 21;
+    
+    
+    output{
+    	
+    	display quartiers{
+    		species quartier;
+    		//species building;
+    	}
+    	
+    	display travel_distance {
+    		chart "Distribution de la distance de trajet des travailleurs" type: histogram background: #lightgrey size: {0.5, 0.5} position: {0, 0}{
+				data "0 - 1 km" value: worker count (each.living_place distance_to each.destination_place < 1000 #m) color: #blue;
+				
+				data "1 - 2 km" value: worker count (each.living_place distance_to each.destination_place >= 1000 #m and each.living_place distance_to each.destination_place < 2000 #m) color: #blue;
+				
+				data "2 - 3 km" value: worker count (each.living_place distance_to each.destination_place >= 2000 #m and each.living_place distance_to each.destination_place < 3000 #m) color: #blue;
+				
+				data "3 - 4 km" value: worker count (each.living_place distance_to each.destination_place >= 3000 #m and each.living_place distance_to each.destination_place < 4000 #m) color: #blue;
+				
+				data "4 - 5 km" value: worker count (each.living_place distance_to each.destination_place >= 4000 #m and each.living_place distance_to each.destination_place < 5000 #m) color: #blue;
+				
+				data "5 - 6 km" value: worker count (each.living_place distance_to each.destination_place >= 5000 #m and each.living_place distance_to each.destination_place < 6000 #m) color: #blue;
+				
+				data "6 - 7 km" value: worker count (each.living_place distance_to each.destination_place >= 6000 #m and each.living_place distance_to each.destination_place < 7000 #m) color: #blue;
+				
+				data "7 - 8 km" value: worker count (each.living_place distance_to each.destination_place >= 7000 #m and each.living_place distance_to each.destination_place < 8000 #m) color: #blue;
+				
+				data "8 - 9 km" value: worker count (each.living_place distance_to each.destination_place >= 8000 #m and each.living_place distance_to each.destination_place < 9000 #m) color: #blue;
+				
+				data "9 - 10 km" value: worker count (each.living_place distance_to each.destination_place >= 9000 #m and each.living_place distance_to each.destination_place < 10000 #m) color: #blue;
+				
+				data "10 - 11 km" value: worker count (each.living_place distance_to each.destination_place >= 10000 #m and each.living_place distance_to each.destination_place < 11000 #m) color: #blue;
+				
+				data "11 - 12 km" value: worker count (each.living_place distance_to each.destination_place >= 11000 #m and each.living_place distance_to each.destination_place < 12000 #m) color: #blue;
+				
+				data "12 - 13 km" value: worker count (each.living_place distance_to each.destination_place >= 12000 #m and each.living_place distance_to each.destination_place < 13000 #m) color: #blue;
+				
+				data "13 - 14 km" value: worker count (each.living_place distance_to each.destination_place >= 13000 #m and each.living_place distance_to each.destination_place < 14000 #m) color: #blue;
+				
+				data "14 - 15 km" value: worker count (each.living_place distance_to each.destination_place >= 14000 #m and each.living_place distance_to each.destination_place < 15000 #m) color: #blue;
+				
+				data "15 - 16 km" value: worker count (each.living_place distance_to each.destination_place >= 15000 #m and each.living_place distance_to each.destination_place < 16000 #m) color: #blue;
+				
+				data "16 - 17 km" value: worker count (each.living_place distance_to each.destination_place >= 16000 #m and each.living_place distance_to each.destination_place < 17000 #m) color: #blue;
+				
+				data "17 - 18 km" value: worker count (each.living_place distance_to each.destination_place >= 17000 #m and each.living_place distance_to each.destination_place < 18000 #m) color: #blue;
+				
+				data "18 - 19 km" value: worker count (each.living_place distance_to each.destination_place >= 18000 #m and each.living_place distance_to each.destination_place < 19000 #m) color: #blue;
+				
+				data "19 - 20 km" value: worker count (each.living_place distance_to each.destination_place >= 19000 #m and each.living_place distance_to each.destination_place < 20000 #m) color: #blue;
+			}
+			
+			chart "Distribution de la distance de trajet des étudiants" type: histogram background: #lightgrey size: {0.5, 0.5} position: {0, 0.5}{
+				data "0 - 1 km" value: student count (each.living_place distance_to each.destination_place < 1000 #m) color: #blue;
+				
+				data "1 - 2 km" value: student count (each.living_place distance_to each.destination_place >= 1000 #m and each.living_place distance_to each.destination_place < 2000 #m) color: #blue;
+				
+				data "2 - 3 km" value: student count (each.living_place distance_to each.destination_place >= 2000 #m and each.living_place distance_to each.destination_place < 3000 #m) color: #blue;
+				
+				data "3 - 4 km" value: student count (each.living_place distance_to each.destination_place >= 3000 #m and each.living_place distance_to each.destination_place < 4000 #m) color: #blue;
+				
+				data "4 - 5 km" value: student count (each.living_place distance_to each.destination_place >= 4000 #m and each.living_place distance_to each.destination_place < 5000 #m) color: #blue;
+				
+				data "5 - 6 km" value: student count (each.living_place distance_to each.destination_place >= 5000 #m and each.living_place distance_to each.destination_place < 6000 #m) color: #blue;
+				
+				data "6 - 7 km" value: student count (each.living_place distance_to each.destination_place >= 6000 #m and each.living_place distance_to each.destination_place < 7000 #m) color: #blue;
+				
+				data "7 - 8 km" value: student count (each.living_place distance_to each.destination_place >= 7000 #m and each.living_place distance_to each.destination_place < 8000 #m) color: #blue;
+				
+				data "8 - 9 km" value: student count (each.living_place distance_to each.destination_place >= 8000 #m and each.living_place distance_to each.destination_place < 9000 #m) color: #blue;
+				
+				data "9 - 10 km" value: student count (each.living_place distance_to each.destination_place >= 9000 #m and each.living_place distance_to each.destination_place < 10000 #m) color: #blue;
+				
+				data "10 - 11 km" value: student count (each.living_place distance_to each.destination_place >= 10000 #m and each.living_place distance_to each.destination_place < 11000 #m) color: #blue;
+				
+				data "11 - 12 km" value: student count (each.living_place distance_to each.destination_place >= 11000 #m and each.living_place distance_to each.destination_place < 12000 #m) color: #blue;
+				
+				data "12 - 13 km" value: student count (each.living_place distance_to each.destination_place >= 12000 #m and each.living_place distance_to each.destination_place < 13000 #m) color: #blue;
+				
+				data "13 - 14 km" value: student count (each.living_place distance_to each.destination_place >= 13000 #m and each.living_place distance_to each.destination_place < 14000 #m) color: #blue;
+				
+				data "14 - 15 km" value: student count (each.living_place distance_to each.destination_place >= 14000 #m and each.living_place distance_to each.destination_place < 15000 #m) color: #blue;
+				
+				data "15 - 16 km" value: student count (each.living_place distance_to each.destination_place >= 15000 #m and each.living_place distance_to each.destination_place < 16000 #m) color: #blue;
+				
+				data "16 - 17 km" value: student count (each.living_place distance_to each.destination_place >= 16000 #m and each.living_place distance_to each.destination_place < 17000 #m) color: #blue;
+				
+				data "17 - 18 km" value: student count (each.living_place distance_to each.destination_place >= 17000 #m and each.living_place distance_to each.destination_place < 18000 #m) color: #blue;
+				
+				data "18 - 19 km" value: student count (each.living_place distance_to each.destination_place >= 18000 #m and each.living_place distance_to each.destination_place < 19000 #m) color: #blue;
+				
+				data "19 - 20 km" value: student count (each.living_place distance_to each.destination_place >= 19000 #m and each.living_place distance_to each.destination_place < 20000 #m) color: #blue;
+			}
+			
+			chart "Distribution de la distance de trajet des leisure" type: histogram background: #lightgrey size: {0.5, 0.5} position: {0.5, 0}{
+				data "0 - 1 km" value: leisure count (each.living_place distance_to each.destination_place < 1000 #m) color: #blue;
+				
+				data "1 - 2 km" value: leisure count (each.living_place distance_to each.destination_place >= 1000 #m and each.living_place distance_to each.destination_place < 2000 #m) color: #blue;
+				
+				data "2 - 3 km" value: leisure count (each.living_place distance_to each.destination_place >= 2000 #m and each.living_place distance_to each.destination_place < 3000 #m) color: #blue;
+				
+				data "3 - 4 km" value: leisure count (each.living_place distance_to each.destination_place >= 3000 #m and each.living_place distance_to each.destination_place < 4000 #m) color: #blue;
+				
+				data "4 - 5 km" value: leisure count (each.living_place distance_to each.destination_place >= 4000 #m and each.living_place distance_to each.destination_place < 5000 #m) color: #blue;
+				
+				data "5 - 6 km" value: leisure count (each.living_place distance_to each.destination_place >= 5000 #m and each.living_place distance_to each.destination_place < 6000 #m) color: #blue;
+				
+				data "6 - 7 km" value: leisure count (each.living_place distance_to each.destination_place >= 6000 #m and each.living_place distance_to each.destination_place < 7000 #m) color: #blue;
+				
+				data "7 - 8 km" value: leisure count (each.living_place distance_to each.destination_place >= 7000 #m and each.living_place distance_to each.destination_place < 8000 #m) color: #blue;
+				
+				data "8 - 9 km" value: leisure count (each.living_place distance_to each.destination_place >= 8000 #m and each.living_place distance_to each.destination_place < 9000 #m) color: #blue;
+				
+				data "9 - 10 km" value: leisure count (each.living_place distance_to each.destination_place >= 9000 #m and each.living_place distance_to each.destination_place < 10000 #m) color: #blue;
+				
+				data "10 - 11 km" value: leisure count (each.living_place distance_to each.destination_place >= 10000 #m and each.living_place distance_to each.destination_place < 11000 #m) color: #blue;
+				
+				data "11 - 12 km" value: leisure count (each.living_place distance_to each.destination_place >= 11000 #m and each.living_place distance_to each.destination_place < 12000 #m) color: #blue;
+				
+				data "12 - 13 km" value: leisure count (each.living_place distance_to each.destination_place >= 12000 #m and each.living_place distance_to each.destination_place < 13000 #m) color: #blue;
+				
+				data "13 - 14 km" value: leisure count (each.living_place distance_to each.destination_place >= 13000 #m and each.living_place distance_to each.destination_place < 14000 #m) color: #blue;
+				
+				data "14 - 15 km" value: leisure count (each.living_place distance_to each.destination_place >= 14000 #m and each.living_place distance_to each.destination_place < 15000 #m) color: #blue;
+				
+				data "15 - 16 km" value: leisure count (each.living_place distance_to each.destination_place >= 15000 #m and each.living_place distance_to each.destination_place < 16000 #m) color: #blue;
+				
+				data "16 - 17 km" value: leisure count (each.living_place distance_to each.destination_place >= 16000 #m and each.living_place distance_to each.destination_place < 17000 #m) color: #blue;
+				
+				data "17 - 18 km" value: leisure count (each.living_place distance_to each.destination_place >= 17000 #m and each.living_place distance_to each.destination_place < 18000 #m) color: #blue;
+				
+				data "18 - 19 km" value: leisure count (each.living_place distance_to each.destination_place >= 18000 #m and each.living_place distance_to each.destination_place < 19000 #m) color: #blue;
+				
+				data "19 - 20 km" value: leisure count (each.living_place distance_to each.destination_place >= 19000 #m and each.living_place distance_to each.destination_place < 20000 #m) color: #blue;
+			}
+			
+			chart "Distance moyenne des trajets en fonction du motif" type: histogram background: #lightgrey size: {0.5, 0.5} position: {0.5, 0.5} {
+    			list<float> travel_distance;
+    		
+    			if(length(worker) != 0){
+	    			loop work over: worker{
+	    				travel_distance << work.living_place distance_to work.destination_place;
+	    			}
+	    			data "Travail" value: sum(travel_distance) / length(worker) color: #blue;
+    			}
+    			
+    		 	if(length(student) != 0){
+	    			travel_distance <- [];
+	    			loop stud over: student{
+	    				travel_distance << stud.living_place distance_to stud.destination_place;
+	    			}
+	    			data "Etude" value: sum(travel_distance) / length(student) color: #blue;
+    			}
+    			
+    			if(length(leisure) != 0){
+	    			travel_distance <- [];
+	    			loop leis over: leisure{
+	    				travel_distance << leis.living_place distance_to leis.destination_place;
+	    			}
+	    			data "Loisir" value: sum(travel_distance) / length(leisure) color: #blue;
+    			}
+
+    			if (length(delivery) != 0){
+	    			travel_distance <- [];
+	    			loop deli over: delivery {
+	    				travel_distance << deli.living_place distance_to deli.restaurant_1;
+	    			}
+	    			data "Distance domicile / restaurant 1" value: sum(travel_distance) / length(delivery) color: #blue;
+	    			
+	    			travel_distance <- [];
+	    			loop deli over: delivery {
+	    				travel_distance << deli.restaurant_1 distance_to deli.residential_place_1;
+	    			}
+	    			data "Distance resto1 / resi1" value: sum(travel_distance) / length(delivery) color: #blue; 
+    			}   		    			
+			}
+    	}
+    	
+    	display hour_travel {
+    		chart "Distribution de l'heure de départ des agents worker" type: histogram background: #lightgrey size: {0.5, 0.5} position: {0, 0}{
+    			loop i from: 0 to: 240 step: 5{
+    				int hour_go <- int(i / 60);
+    				data string(i) value: worker count (each.go_out_hour = int(i / 60) + 6 and each.go_out_min = int((i - (each.go_out_hour - 6) * 60) / 5) * 5) color: #blue;
+    			}
+    		}
+    		
+    		chart "Distribution de l'heure de départ des agents student" type: histogram background: #lightgrey size: {0.5, 0.5} position: {0.5, 0.5}{
+    			loop i from: 0 to: 180 step: 5{
+    				int hour_go <- int(i / 60);
+    				data string(i) value: student count (each.go_out_hour = int(i / 60) + 6 and each.go_out_min = int((i - (each.go_out_hour - 6) * 60) / 5) * 5) color: #blue;
+    			}
+    		}	
+    	} 	
+	}
+}
+
+
+
+
+
+

4_Bicycle network Marseille travel agents.gaml

+/**
+* Name: Deplacementdevelos
+* Based on the internal empty template. 
+* Author: nathancoisne
+* Tags: 
+*/
+
+
+model Deplacementdevelos
+
+/* 
+ * This model generates the daily trips (non-participatory trip = the shortest path) of a synthetic population of cyclists in Marseille
+ * The living place, destination place and hour of leaving / coming home are already defined for each agent in the synthetic population
+ * 
+ * 
+ */
+
+global {
+	
+	bool test_population <- false;
+	
+	file bound <- shape_file("../includes/Marseille/boundary_Marseille.shp");
+	
+	file shape_file_buildings;
+                            
+    file shape_file_roads;
+      
+	bool workers <- false;
+    bool students <- true;
+    bool leisures <- false;
+    bool deliveries <- false;
+    	           
+    file worker_csv;
+    
+    file student_csv;
+    
+    file leisure_csv;
+    
+    file delivery_midi_csv;
+    
+    file delivery_soir_csv;
+        
+    geometry shape <- envelope(bound);
+    
+   	int time_step <- 5; // time step in seconds
+   	
+    float step <- time_step #s;
+    
+    date starting_date <- date(2021, 5, 27, 5, 30);
+        
+    float bicycle_speed <- 10 #km / #h;
+	
+    graph the_graph;
+        
+    map<road,float> road_weights;
+    
+    reflex save_trips when: current_date.hour = 23 and current_date.minute = 55{
+    	
+    	list<position_agent> list_of_positions_aller_workers;
+    	list<position_agent> list_of_positions_retour_workers;
+    	
+    	ask worker {
+    		loop pos over: list_of_positions_aller{
+    			list_of_positions_aller_workers << pos;
+    		}
+    		
+    		loop pos over: list_of_positions_retour{
+    			list_of_positions_retour_workers << pos;
+    		}
+    	}
+    	
+    	save list_of_positions_aller_workers to: "../results/Marseille/worker/aller_worker/positions_aller_worker.shp" type: shp 
+			attributes: ["time"::date_of_presence, "agent"::name_of_agent] crs: "2154";
+			
+		save list_of_positions_retour_workers to: "../results/Marseille/worker/retour_worker/positions_retour_worker.shp" type: shp 
+			attributes: ["time"::date_of_presence, "agent"::name_of_agent] crs: "2154";
+			
+		
+		list<position_agent> list_of_positions_aller_students;
+    	list<position_agent> list_of_positions_retour_students;
+    	
+    	ask student {
+    		loop pos over: list_of_positions_aller{
+    			list_of_positions_aller_students << pos;
+    		}
+    		
+    		loop pos over: list_of_positions_retour{
+    			list_of_positions_retour_students << pos;
+    		}
+    	}
+    	
+    	save list_of_positions_aller_students to: "../results/Marseille/student/aller_student/positions_aller_student.shp" type: shp 
+			attributes: ["time"::date_of_presence, "agent"::name_of_agent] crs: "2154";
+			
+		save list_of_positions_retour_students to: "../results/Marseille/student/retour_student/positions_retour_student.shp" type: shp 
+			attributes: ["time"::date_of_presence, "agent"::name_of_agent] crs: "2154";
+			
+			
+		list<position_agent> list_of_positions_aller_leisures;
+    	list<position_agent> list_of_positions_retour_leisures;
+    	
+    	ask leisure {
+    		loop pos over: list_of_positions_aller{
+    			list_of_positions_aller_leisures << pos;
+    		}
+    		
+    		loop pos over: list_of_positions_retour{
+    			list_of_positions_retour_leisures << pos;
+    		}
+    	}
+    	
+    	save list_of_positions_aller_leisures to: "../results/Marseille/leisure/aller_leisure/positions_aller_leisure.shp" type: shp 
+			attributes: ["time"::date_of_presence, "agent"::name_of_agent] crs: "2154";
+			
+		save list_of_positions_retour_leisures to: "../results/Marseille/leisure/retour_leisure/positions_retour_leisure.shp" type: shp 
+			attributes: ["time"::date_of_presence, "agent"::name_of_agent] crs: "2154";
+    	
+    }
+
+	reflex stop_simulation when: current_date.hour = 23 and current_date.minute = 55 {
+		do pause;
+	}
+
+
+    init{    	
+	   	shape_file_buildings <- shape_file("../includes/Marseille/buildings.shp"); // building contenant les data osm, ign, pois
+	   	shape_file_roads <- shape_file("../includes/Marseille/SHP Marseille/roads_cyclists.shp");
+	   	
+	   	if (test_population){
+			write("Marseille : population test");
+			worker_csv <- csv_file( "../results/Marseille/synthetic_population_test/worker.csv",true);
+		
+		    student_csv <- csv_file( "../results/Marseille/synthetic_population_test/student.csv",true);
+		    
+		    leisure_csv <- csv_file( "../results/Marseille/synthetic_population_test/leisure.csv",true);
+		    
+		    delivery_midi_csv <- csv_file( "../results/Marseille/synthetic_population_test/delivery_midi.csv",true);
+		    
+		    delivery_soir_csv <- csv_file( "../results/Marseille/synthetic_population_test/delivery_soir.csv",true);
+		}
+	   	else{
+			write("Marseille : population reelle");
+			worker_csv <- csv_file( "../results/Marseille/synthetic_population/worker.csv",true);
+		
+		    student_csv <- csv_file( "../results/Marseille/synthetic_population/student.csv",true);
+		    
+		    leisure_csv <- csv_file( "../results/Marseille/synthetic_population/leisure.csv",true);
+		    
+		    delivery_midi_csv <- csv_file( "../results/Marseille/synthetic_population/delivery_midi.csv",true);
+		    
+		    delivery_soir_csv <- csv_file( "../results/Marseille/synthetic_population/delivery_soir.csv",true);
+		}
+	
+ 
+    	create building from: shape_file_buildings {
+    		types <- types_str split_with ",";
+    	}
+       	    	
+    	create road from: shape_file_roads with: [id::int(get("id")), fclass::read ("fclass"), speed_coeff::float(get("s_coeff")), maxspeed::int(get("maxspeed"))];
+    	
+    	road_weights <- road as_map (each::each.shape.perimeter / each.speed_coeff); // speed_coeff depends on the type of road
+    	the_graph <- as_edge_graph(road);
+		the_graph <- the_graph with_weights road_weights;    	
+    	
+    	map<int,building> building_map <- building as_map (each.id::each);
+
+		if workers{
+	    	write("instanciate workers");
+	    	
+	    	create worker from: worker_csv with:[go_out_hour::int(get("go_out_hour")), go_out_min::int(get("go_out_min")), go_home_hour::int(get("go_home_hour")), go_home_min::int(get("go_home_min")),
+	    										id_living_place::int(get("living_place.id")), id_destination_place::int(get("destination_place.id"))]{
+	    																
+				date_of_leaving_home <- date(2021, 5, 27, go_out_hour, go_out_min);
+				date_of_leaving_dest <- date(2021, 5, 27, go_home_hour, go_home_min);
+																				   	
+	    		living_place <- building_map[id_living_place];																				 
+	    		location <- any_location_in (living_place);
+	    		destination_place <- building_map[id_destination_place];		
+				speed <- bicycle_speed;
+					
+	        	objective <- "resting";
+	        	is_on_road <- false;
+	    	}
+    	}
+    	
+    	if students{
+	    	write("instanciate students");
+	    	
+	    	create student from: student_csv with:[go_out_hour::int(get("go_out_hour")), go_out_min::int(get("go_out_min")), go_home_hour::int(get("go_home_hour")), go_home_min::int(get("go_home_min")),
+	    										id_living_place::int(get("living_place.id")), id_destination_place::int(get("destination_place.id"))]{
+	    					
+				date_of_leaving_home <- date(2021, 5, 27, go_out_hour, go_out_min);
+				date_of_leaving_dest <- date(2021, 5, 27, go_home_hour, go_home_min);
+				
+	    		living_place <- building_map[id_living_place];																				 
+	    		location <- any_location_in (living_place);
+	    		destination_place <- building_map[id_destination_place];		
+				speed <- bicycle_speed;
+					    		
+	        	objective <- "resting";
+	        	is_on_road <- false;
+	  	    }
+	    }
+
+		if leisures{
+	    	write("instanciate leisures");
+	    	create leisure from: leisure_csv with:[go_out_hour::int(get("go_out_hour")), go_out_min::int(get("go_out_min")), go_home_hour::int(get("go_home_hour")), go_home_min::int(get("go_home_min")),
+	    										id_living_place::int(get("living_place.id")), id_destination_place::int(get("destination_place.id"))]{
+	    														
+	    		date_of_leaving_home <- date(2021, 5, 27, go_out_hour, go_out_min);
+	    		date_of_leaving_dest <- date(2021, 5, 27, go_home_hour, go_home_min);
+	    										   	
+	    		living_place <- building_map[id_living_place];																				 
+	    		location <- any_location_in (living_place);
+	    		destination_place <- building_map[id_destination_place];		
+				speed <- bicycle_speed;
+	    			    		
+	        	objective <- "resting";
+	        	is_on_road <- false;
+	    	}
+	    }
+    }          
+}
+
+species building {
+	int id;
+    string type; 
+    string types_str;
+    list<string> types;
+    float height;
+    rgb color <- #gray;
+        
+    aspect base {
+    draw shape color: color ;
+    }
+}
+
+species road  {
+	int id;
+	
+	string fclass;
+	
+	int maxspeed;
+	
+	float speed_coeff;
+	
+	bool starting_road_aller;
+	
+	bool starting_road_retour;
+	
+	rgb color <- #orange; 
+		
+    aspect base {
+    draw shape color: color width: 1;
+    }
+}
+
+species people skills: [moving] {
+	rgb color;
+	
+	int id_living_place;
+	int id_destination_place;
+	building living_place <- nil;
+	building destination_place <- nil;
+	
+	path path_followed;
+		
+	int go_out_hour;
+	int go_out_min;
+	date date_of_leaving_home;
+	
+	date date_of_arriving_dest;
+	
+	path path_aller;
+	list<road> list_of_roads_aller;
+	list<position_agent> list_of_positions_aller;
+	
+	int go_home_hour;
+	int go_home_min;
+	date date_of_leaving_dest;
+	
+	date date_of_arriving_home;
+	
+	path path_retour;
+	list<position_agent> list_of_positions_retour;
+	list<road> list_of_roads_retour;
+		
+	bool is_on_road;
+		
+	string objective; 
+	
+	point the_target <- nil;
+	
+	reflex time_to_leave when: current_date.hour = go_out_hour and current_date.minute = go_out_min and objective = "resting"{
+		objective <- "working";
+		the_target <- any_location_in (destination_place);
+
+		path_aller <- path_between(the_graph, location, the_target);
+						
+		if path_aller != nil{
+			loop rd over: path_aller.edges{
+				list_of_roads_aller << road(int(rd));
+			}
+		}
+		
+		list_of_roads_aller[0].starting_road_aller <- true;
+			
+		is_on_road <- true;
+	}
+	
+	reflex time_to_go_home when: current_date.hour = go_home_hour and current_date.minute = go_home_min and objective = "working"{
+		objective <- "resting";
+		the_target <- any_location_in (living_place);
+		
+		
+		path_retour <- path_between(the_graph, location, the_target);
+				
+		if path_retour != nil{
+			loop rd over: path_retour.edges{
+				list_of_roads_retour << road(int(rd));
+			}
+		}	
+		
+		list_of_roads_retour[0].starting_road_retour <- true;
+		
+		is_on_road <- true;
+	}
+	
+	reflex move when: the_target != nil {
+		point current_location <- location;
+		
+		create position_agent{
+			location <- current_location;
+			date_of_presence <- current_date;
+			name_of_agent <- string(myself);
+		}
+		
+		if objective = "working"{
+			list_of_positions_aller << last(position_agent);
+		}
+		if objective = "resting"{
+			list_of_positions_retour << last(position_agent);
+		}
+		
+		path_followed <- goto(target: the_target, on: the_graph, return_path: true);
+		
+		if the_target = location {
+	    	the_target <- nil;
+	    	is_on_road <- false;
+	    	if (objective = "working"){
+		    	date_of_arriving_dest <- date(2021, 5, 27, current_date.hour, current_date.minute, current_date.second);
+		    	
+	    	}
+	    	if (objective = "resting"){
+		    	date_of_arriving_home <- date(2021, 5, 27, current_date.hour, current_date.minute, current_date.second);
+	    	}
+	    }
+		
+    }
+    
+	aspect base{
+		draw circle(10) color: color border: #black;
+	}
+}
+
+species worker parent: people {
+	rgb color <- #black;
+}
+
+species student parent: people{
+	rgb color <- #green;
+}
+
+species leisure parent: people{
+	rgb color <- #pink;
+}
+
+species position_agent {
+	date date_of_presence;
+	string name_of_agent;
+}
+
+
+experiment bike_traffic type: gui {
+       
+    output{
+    	display city_display type: opengl{
+    		//species building aspect: base;
+    		species road aspect: base;
+    		species worker aspect: base;
+    		species student aspect: base;
+    		species leisure aspect: base;
+    	}
+
+		display nb_persons_on_road refresh: every(2#cycles){
+    		chart "Nombre d'agents sur les routes" type: series {
+				data "Nombre d'agents sur les routes" value: worker count (each.is_on_road = true) + student count (each.is_on_road = true) + leisure count (each.is_on_road = true);
+			}
+    	}
+    	
+    	/*
+    	display pollution type: opengl{
+    		species pollution_cell aspect: pollution transparency: 0;
+    	}
+    
+    	* 
+
+    	display hour_travel {
+    		chart "Distribution de l'heure de départ des agents worker" type: histogram background: #lightgrey{
+    			loop i from: 0 to: 240 step: 5{
+    				int hour_go <- int(i / 60);
+    				data string(i) value: worker count (each.go_out_hour = int(i / 60) + 6 and each.go_out_min = int((i - (each.go_out_hour - 6) * 60) / 5) * 5) color: #blue;
+    			}
+    		}
+
+    	}
+    	* 
+    	*/
+
+		monitor "Date" value: current_date ;	
+	}
+	
+	
+	
+	
+	
+	
+	
+}
\ No newline at end of file

5_Synthetic measures from agents' travels.gaml

+/**
+* Name: Deplacementdevelos
+* Based on the internal empty template. 
+* Author: nathancoisne
+* Tags: 
+*/
+
+
+model Deplacementdevelos
+
+/* 
+ * This model generates pollution measures from the trips generated by 'travel agents'. Pollution data are extracted from a hourly pollution model from AtmoSud (date: 27/05/2021)
+ * 
+ */
+
+global {
+		
+	file bound <- shape_file("../includes/Marseille/boundary_Marseille.shp");
+		  	
+	bool workers <- false; // une seule population en 'true' à la fois pour le stockage des mesures par type d'agent
+    bool students <- false;
+    bool leisures <- true;
+    
+    bool measure_NO2 <- true;
+    bool measure_O3 <- true;
+    bool measure_PM10 <- true;
+    bool measure_PM25 <- true;
+        
+    geometry shape <- envelope(bound);
+    
+    // coordonees des rasters de pollution
+    float min_x <- 889462.50000;
+    float min_y <- 6236287.50000;
+    float max_x <- 905062.50000;
+    float max_y <- 6254412.50000;
+    
+    int compteur_ID <- 0;
+   	
+    init{ 
+  		   	
+		// mise en mémoire du cube de pollution NO2, O3, PM10, PM25 (24 rasters horaires)
+		list<matrix<float>> NO2_rasters;
+		list<matrix<float>> O3_rasters;
+		list<matrix<float>> PM10_rasters;
+		list<matrix<float>> PM25_rasters;
+		
+		if measure_NO2 {
+			// raster heure 0 manquant pour la journée du 27
+			matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+			NO2_rasters << pollution_raster_float;
+			
+			loop i from: 1 to: 23{
+				
+				string path_grid;
+				if (i < 10){
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_NO2_202105270' + i + '_202105270' + i + '.tif';
+				}
+				else{
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_NO2_20210527' + i + '_20210527' + i + '.tif';
+				}
+				
+				matrix my_data <- grid_file(path_grid) as_matrix({624, 725});
+		        
+		        matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+					
+				loop j from: 0 to: 623{
+					loop k from: 0 to: 724{
+						
+						pollution_raster_float[j, k] <- float (my_data[j, k] get("grid_value"));
+						
+					}
+				}
+				write ('raster NO2 ' + i + ' chargé');
+				write('max: ' + max(pollution_raster_float));
+				
+				NO2_rasters << pollution_raster_float;
+			}
+		}
+		
+		if measure_O3 {
+			
+			matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+			O3_rasters << pollution_raster_float;
+			
+			loop i from: 1 to: 23{
+				
+				string path_grid;
+				if (i < 10){
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_O3_202105270' + i + '_202105270' + i + '.tif';
+				}
+				else{
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_O3_20210527' + i + '_20210527' + i + '.tif';
+				}
+				
+				matrix my_data <- grid_file(path_grid) as_matrix({624, 725});
+		        
+		        matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+					
+				loop j from: 0 to: 623{
+					loop k from: 0 to: 724{
+						
+						pollution_raster_float[j, k] <- float (my_data[j, k] get("grid_value"));
+						
+					}
+				}
+				write ('raster O3 ' + i + ' chargé');
+				write('max: ' + max(pollution_raster_float));
+				
+				O3_rasters << pollution_raster_float;
+			}
+		}
+		
+		if measure_PM10 {
+			
+			matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+			PM10_rasters << pollution_raster_float;
+			
+			loop i from: 1 to: 23{
+				
+				string path_grid;
+				if (i < 10){
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_PM10_202105270' + i + '_202105270' + i + '.tif';
+				}
+				else{
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_PM10_20210527' + i + '_20210527' + i + '.tif';
+				}
+				
+				matrix my_data <- grid_file(path_grid) as_matrix({624, 725});
+		        
+		        matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+					
+				loop j from: 0 to: 623{
+					loop k from: 0 to: 724{
+						
+						pollution_raster_float[j, k] <- float (my_data[j, k] get("grid_value"));
+						
+					}
+				}
+				write ('raster PM10 ' + i + ' chargé');
+				write('max: ' + max(pollution_raster_float));
+				
+				PM10_rasters << pollution_raster_float;
+			}
+		}
+		
+		if measure_PM25 {
+			
+			matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+			PM25_rasters << pollution_raster_float;
+			
+			loop i from: 1 to: 23{
+				
+				string path_grid;
+				if (i < 10){
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_PM25_202105270' + i + '_202105270' + i + '.tif';
+				}
+				else{
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_PM25_20210527' + i + '_20210527' + i + '.tif';
+				}
+				
+				matrix my_data <- grid_file(path_grid) as_matrix({624, 725});
+		        
+		        matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+					
+				loop j from: 0 to: 623{
+					loop k from: 0 to: 724{
+						
+						pollution_raster_float[j, k] <- float (my_data[j, k] get("grid_value"));
+						
+					}
+				}
+				write ('raster PM25 ' + i + ' chargé');
+				write('max: ' + max(pollution_raster_float));
+				
+				PM25_rasters << pollution_raster_float;
+			}
+		}
+		
+		
+		// generation des donnees de pollution pour chaque type d'agent
+		
+		if workers{
+	    	write("generating workers measures");
+	    		    	
+	    	create measure from: shape_file("../results/Marseille/worker/aller_worker/positions_aller_worker.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    		
+	    		int worker_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(worker_number);
+				if (worker_number > 3000){do die;} // on calcule les mesures faites pour une majorité d'agent. le reste servira de test aux prédictions faites à partir de ces mesures
+
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+				  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/worker/workers_measures.csv" type:"csv" rewrite: false;
+	    	}
+	    	
+	    	ask measure{do die;}
+	    	
+	    	create measure from: shape_file("../results/Marseille/worker/retour_worker/positions_retour_worker.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    		
+	    		int worker_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(worker_number);
+				if (worker_number > 3000){do die;} // on calcule les mesures faites pour une majorité d'agent. le reste servira de test aux prédictions faites à partir de ces mesures
+
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/worker/workers_measures.csv" type:"csv" rewrite: false;
+	    	}
+	    	ask measure{do die;}
+	    	
+	    		
+	    }	    	
+	    
+    	   		
+    	if students{
+	    	write("generating students measures");
+	    		    	
+	    	create measure from: shape_file("../results/Marseille/student/aller_student/positions_aller_student.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    		
+	    		int student_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(student_number);
+				if (student_number > 1800){do die;} // on calcule les mesures faites pour une majorité d'agent. le reste servira de test aux prédictions faites à partir de ces mesures
+				
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+				  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/student/students_measures.csv" type:"csv" rewrite: false;
+	    	}
+	    	
+	    	ask measure{do die;}
+	    	
+	    	create measure from: shape_file("../results/Marseille/student/retour_student/positions_retour_student.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    			    		
+	    		int student_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(student_number);
+				if (student_number > 1800){do die;} // on calcule les mesures faites pour une majorité d'agent. le reste servira de test aux prédictions faites à partir de ces mesures
+	    		
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/student/students_measures.csv" type:"csv" rewrite: false;
+	    	}
+	    	ask measure{do die;}
+	    		
+	    }
+	    
+	    if leisures{
+	    	write("generating leisures measures");
+	    		    	
+	    	create measure from: shape_file("../results/Marseille/leisure/aller_leisure/positions_aller_leisure.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    		
+	    		int leisure_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(leisure_number);
+				if (leisure_number > 6200){do die;} // on calcule les mesures faites pour une majorité d'agent. le reste servira de test aux prédictions faites à partir de ces mesures
+
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+				  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/leisure/leisures_measures.csv" type:"csv" rewrite: false;
+	    	}
+	    	
+	    	ask measure{do die;}
+	    	
+	    	create measure from: shape_file("../results/Marseille/leisure/retour_leisure/positions_retour_leisure.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    		
+	    		int leisure_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(leisure_number);
+				if (leisure_number > 6200){do die;} // on calcule les mesures faites pour une majorité d'agent. le reste servira de test aux prédictions faites à partir de ces mesures
+	    		
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/leisure/leisures_measures.csv" type:"csv" rewrite: false;
+	    	}
+	    	ask measure{do die;}
+	    		
+	    }		    	
+    }          
+}
+
+species measure {
+	int ID;
+	string name_of_agent;
+	float longitude min: min_x max: max_x - 0.1;
+	float latitude min: min_y max: max_y - 0.1;
+	date time_of_measure;
+	float NO2_concentration;
+	float O3_concentration;
+	float PM10_concentration;
+	float PM25_concentration;
+}
+
+grid pollution_cell file: grid_file('../includes/Marseille/pollution_model/raster_dep13_NO2_2021052701_2021052701.tif'){
+}
+
+
+experiment bike_traffic type: gui {
+
+}
\ No newline at end of file

6_Environment for measures.gaml

+/**
+* Name: Deplacementdevelos
+* Based on the internal empty template. 
+* Author: nathancoisne
+* Tags: 
+*/
+
+
+model Deplacementdevelos
+
+/* 
+ * This model generates the environment close to the measures point
+ * 
+ */
+
+global {
+	bool workers <- false;
+    bool students <- true;
+    bool leisures <- false;
+	
+	file measures_workers_csv;
+    file measures_students_csv;
+    file measures_leisures_csv;
+    
+    file bound <- shape_file("../includes/Marseille/boundary_Marseille.shp");
+    geometry shape <- envelope(bound);
+    
+    file shape_file_buildings <- shape_file("../includes/Marseille/buildings_ign.shp");
+    
+    file shape_file_vegetation <- shape_file("../includes/Marseille/environment/ZONE_DE_VEGETATION.shp");
+    
+    file roads_ign <- shape_file("../includes/Marseille/environment/roads_ign.shp");
+    
+    file primary_roads <- shape_file("../includes/Marseille/environment/primary_roads.shp");
+
+
+    init{    	
+
+    	create building from: shape_file_buildings with: [height::float(read("HAUTEUR"))];
+    	
+    	create road from: roads_ign;
+    	list<road> rd_1_voie <- road where (each.NB_VOIES = 1);
+    	list<road> rd_2_voie <- road where (each.NB_VOIES = 2);
+    	list<road> rd_3_voie <- road where (each.NB_VOIES = 3);
+    	list<road> rd_4_voie <- road where (each.NB_VOIES = 4);
+    	list<road> rd_5_voie <- road where (each.NB_VOIES = 5);
+    	list<road> rd_6_voie <- road where (each.NB_VOIES = 6);
+    	
+    	list<road> list_rd_0_4_width <- road where (each.LARGEUR <= 4.0);
+    	list<road> list_rd_4_6_width <- road where (each.LARGEUR > 4.0 and each.LARGEUR <= 6.0);
+    	list<road> list_rd_6_8_width <- road where (each.LARGEUR > 6.0 and each.LARGEUR <= 8.0);
+    	list<road> list_rd_8_max_width <- road where (each.LARGEUR > 8.0);
+    	
+    	create primary_road from: primary_roads;
+    	
+    	create vegetation from: shape_file_vegetation;
+    	
+    	list<vegetation> list_bois <- vegetation where (each.NATURE = 'Bois');
+    	list<vegetation> list_foret <- vegetation where (each.NATURE = 'Forêt fermée de conifères' or each.NATURE = 'Forêt fermée de feuillus'
+    												     or each.NATURE = 'Forêt fermée mixte' or each.NATURE = 'Forêt ouverte');
+    	list<vegetation> list_haie <- vegetation where (each.NATURE = 'Haie');
+    	
+    	if(workers){
+    		measures_workers_csv <- csv_file("../results/Marseille/worker/workers_measures.csv",true);
+    		
+	    	create measure from: measures_workers_csv{
+	    		write(self.ID);
+	    		location <- point(to_GAMA_CRS({self.longitude, self.latitude}, 'EPSG:2154'));
+	    		geometry disque <- circle(50, location);
+	    		
+	    		create environment {
+	    			ID <- myself.ID;
+	    			name_of_agent <- myself.name_of_agent;
+	    			longitude <- myself.longitude;
+	    			latitude <- myself.latitude;
+	    			
+	    			loop build over: building overlapping disque{
+						float surface_inter <- (build inter disque).area;
+						buildings_volume <- buildings_volume + build.height * surface_inter;
+					}
+					
+					loop rd over: rd_1_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_1 <- voie_1 + perimeter_inter;
+					}
+					
+					loop rd over: rd_2_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_2 <- voie_2 + perimeter_inter;
+					}
+					
+					loop rd over: rd_3_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_3 <- voie_3 + perimeter_inter;
+					}
+					
+					loop rd over: rd_4_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_4 <- voie_4 + perimeter_inter;
+					}
+					
+					loop rd over: rd_5_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_5 <- voie_5 + perimeter_inter;
+					}
+					
+					loop rd over: rd_6_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_6 <- voie_6 + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_0_4_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_0_4_width <- road_0_4_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_4_6_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_4_6_width <- road_4_6_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_6_8_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_6_8_width <- road_6_8_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_8_max_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_8_max_width <- road_8_max_width + perimeter_inter;
+					}
+					
+					loop veg over: list_bois overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						bois <- bois + area_inter;
+					}
+					
+					loop veg over: list_foret overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						foret <- foret + area_inter;
+					}
+					
+					loop veg over: list_haie overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						haie <- haie + area_inter;
+					}
+					
+					distance_to_main_road <- myself distance_to (primary_road closest_to myself);
+					
+	    		}
+	    	}
+	    	
+	    	ask environment {
+		    			save [ID, name_of_agent, longitude, latitude, buildings_volume, voie_1, voie_2, voie_3, voie_4, voie_5, voie_6, road_0_4_width, road_4_6_width, road_6_8_width, road_8_max_width, distance_to_main_road, bois, foret, haie]
+	    				to: "../results/Marseille/worker/workers_measures_environment.csv" type:"csv" rewrite: false;   
+		    }   
+		}
+    	
+    	if(students){
+    		measures_students_csv <- csv_file("../results/Marseille/student/students_measures.csv",true);
+    		
+	    	create measure from: measures_students_csv{
+	    		write(self.ID);
+	    		location <- point(to_GAMA_CRS({self.longitude, self.latitude}, "EPSG:2154"));
+	    		geometry disque <- circle(50, location);
+	    		create environment {
+	    			ID <- myself.ID;
+	    			name_of_agent <- myself.name_of_agent;
+	    			longitude <- myself.longitude;
+	    			latitude <- myself.latitude;
+	    			
+	    			loop build over: building overlapping disque{
+						float surface_inter <- (build inter disque).area;
+						buildings_volume <- buildings_volume + build.height * surface_inter;
+					}
+					
+					loop rd over: rd_1_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_1 <- voie_1 + perimeter_inter;
+					}
+					
+					loop rd over: rd_2_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_2 <- voie_2 + perimeter_inter;
+					}
+					
+					loop rd over: rd_3_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_3 <- voie_3 + perimeter_inter;
+					}
+					
+					loop rd over: rd_4_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_4 <- voie_4 + perimeter_inter;
+					}
+					
+					loop rd over: rd_5_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_5 <- voie_5 + perimeter_inter;
+					}
+					
+					loop rd over: rd_6_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_6 <- voie_6 + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_0_4_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_0_4_width <- road_0_4_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_4_6_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_4_6_width <- road_4_6_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_6_8_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_6_8_width <- road_6_8_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_8_max_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_8_max_width <- road_8_max_width + perimeter_inter;
+					}
+					
+					loop veg over: list_bois overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						bois <- bois + area_inter;
+					}
+					
+					loop veg over: list_foret overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						foret <- foret + area_inter;
+					}
+					
+					loop veg over: list_haie overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						haie <- haie + area_inter;
+					}
+					
+					distance_to_main_road <- myself distance_to (primary_road closest_to myself);
+					
+	    		}
+	    	}
+	    	
+	    	ask environment {
+		    			save [ID, name_of_agent, longitude, latitude, buildings_volume, voie_1, voie_2, voie_3, voie_4, voie_5, voie_6, road_0_4_width, road_4_6_width, road_6_8_width, road_8_max_width, distance_to_main_road, bois, foret, haie]
+	    				to: "../results/Marseille/student/students_measures_environment.csv" type:"csv" rewrite: false;   
+		    }   
+		}
+		
+		if(leisures){
+    		measures_leisures_csv <- csv_file("../results/Marseille/leisure/leisures_measures.csv",true);
+    		
+	    	create measure from: measures_leisures_csv{
+	    		write(self.ID);
+	    		location <- point(to_GAMA_CRS({self.longitude, self.latitude}, 'EPSG:2154'));
+	    		geometry disque <- circle(50, location);
+	    		
+	    		create environment {
+	    			ID <- myself.ID;
+	    			name_of_agent <- myself.name_of_agent;
+	    			longitude <- myself.longitude;
+	    			latitude <- myself.latitude;
+	    			
+	    			loop build over: building overlapping disque{
+						float surface_inter <- (build inter disque).area;
+						buildings_volume <- buildings_volume + build.height * surface_inter;
+					}
+					
+					loop rd over: rd_1_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_1 <- voie_1 + perimeter_inter;
+					}
+					
+					loop rd over: rd_2_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_2 <- voie_2 + perimeter_inter;
+					}
+					
+					loop rd over: rd_3_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_3 <- voie_3 + perimeter_inter;
+					}
+					
+					loop rd over: rd_4_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_4 <- voie_4 + perimeter_inter;
+					}
+					
+					loop rd over: rd_5_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_5 <- voie_5 + perimeter_inter;
+					}
+					
+					loop rd over: rd_6_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_6 <- voie_6 + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_0_4_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_0_4_width <- road_0_4_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_4_6_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_4_6_width <- road_4_6_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_6_8_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_6_8_width <- road_6_8_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_8_max_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_8_max_width <- road_8_max_width + perimeter_inter;
+					}
+					
+					loop veg over: list_bois overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						bois <- bois + area_inter;
+					}
+					
+					loop veg over: list_foret overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						foret <- foret + area_inter;
+					}
+					
+					loop veg over: list_haie overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						haie <- haie + area_inter;
+					}
+					
+					distance_to_main_road <- myself distance_to (primary_road closest_to myself);
+					
+	    		}
+	    	}
+	    	
+	    	ask environment {
+		    			save [ID, name_of_agent, longitude, latitude, buildings_volume, voie_1, voie_2, voie_3, voie_4, voie_5, voie_6, road_0_4_width, road_4_6_width, road_6_8_width, road_8_max_width, distance_to_main_road, bois, foret, haie]
+	    				to: "../results/Marseille/leisure/leisures_measures_environment.csv" type:"csv" rewrite: false;   
+		    }   
+		}
+	}
+
+}
+
+species measure { // seulement besoin de la localisation de la mesure, etant donne que l'environnement est suppose stationnaire
+	int ID;
+	string name_of_agent;
+	float longitude;
+	float latitude;
+}
+
+species environment{ // flotants correspondant à l'environnement dans un disque de 50m de rayon centré sur la mesure
+	int ID; // ID égal à celui de la mesure
+	string name_of_agent;
+	float longitude;
+	float latitude;
+	
+	float road_0_4_width; // longueur de route dans le disque ayant comme attribut : largeur <= 4 
+	float road_4_6_width; // longueur de route dans le disque ayant comme attribut : 4 < largeur <= 6 
+	float road_6_8_width; // longueur de route dans le disque ayant comme attribut : 6 < largeur <= 8 
+	float road_8_max_width; // longueur de route dans le disque ayant comme attribut : 8 < largeur
+	float voie_1;
+	float voie_2;
+	float voie_3;
+	float voie_4;
+	float voie_5;
+	float voie_6;
+	float buildings_volume;
+	float distance_to_main_road;
+	float bois; //surface de bois dans le disque
+	float foret;
+	float haie;
+}
+
+species road {
+	float LARGEUR;
+	int NB_VOIES;
+}
+
+species primary_road{
+}
+
+species building {
+	int id;
+    string type; 
+    string types_str;
+    list<string> types;
+    float height;        
+}
+
+species vegetation {
+	string NATURE; //Bois, Forêt fermée de conifères, Forêt fermée de feuillus, Forêt fermée mixte, Forêt ouverte, Haie, Lande ligneuse, Verger, Vigne (types BD TOPO)
+}
+
+
+
+experiment bike_traffic type: gui {
+}
\ No newline at end of file

7_Measures to predict.gaml

+/**
+* Name: Deplacementdevelos
+* Based on the internal empty template. 
+* Author: nathancoisne
+* Tags: 
+*/
+
+
+model Deplacementdevelos
+
+/* 
+ * This model generates pollution measures from the last quarter of trips generated by 'travel agents'. Pollution data are extracted from a hourly pollution model from AtmoSud (date: 27/05/2021)
+ * These measures will be compared to predicted measures
+ * 
+ */
+
+global {
+		
+	file bound <- shape_file("../includes/Marseille/boundary_Marseille.shp");
+		  	
+	bool workers <- false; // une seule population en 'true' à la fois pour le stockage des mesures par type d'agent
+    bool students <- true;
+    bool leisures <- false;
+    
+    bool measure_NO2 <- true;
+    bool measure_O3 <- true;
+    bool measure_PM10 <- true;
+    bool measure_PM25 <- true;
+        
+    geometry shape <- envelope(bound);
+    
+    // coordonees des rasters de pollution
+    float min_x <- 889462.50000;
+    float min_y <- 6236287.50000;
+    float max_x <- 905062.50000;
+    float max_y <- 6254412.50000;
+    
+    int compteur_ID <- 0;
+   	
+    init{ 
+  		   	
+		// mise en mémoire du cube de pollution NO2, O3, PM10, PM25 (24 rasters horaires)
+		list<matrix<float>> NO2_rasters;
+		list<matrix<float>> O3_rasters;
+		list<matrix<float>> PM10_rasters;
+		list<matrix<float>> PM25_rasters;
+		
+		if measure_NO2 {
+			// raster heure 0 manquant pour la journée du 27
+			matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+			NO2_rasters << pollution_raster_float;
+			
+			loop i from: 1 to: 23{
+				
+				string path_grid;
+				if (i < 10){
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_NO2_202105270' + i + '_202105270' + i + '.tif';
+				}
+				else{
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_NO2_20210527' + i + '_20210527' + i + '.tif';
+				}
+				
+				matrix my_data <- grid_file(path_grid) as_matrix({624, 725});
+		        
+		        matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+					
+				loop j from: 0 to: 623{
+					loop k from: 0 to: 724{
+						
+						pollution_raster_float[j, k] <- float (my_data[j, k] get("grid_value"));
+						
+					}
+				}
+				write ('raster NO2 ' + i + ' chargé');
+				write('max: ' + max(pollution_raster_float));
+				
+				NO2_rasters << pollution_raster_float;
+			}
+		}
+		
+		if measure_O3 {
+			
+			matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+			O3_rasters << pollution_raster_float;
+			
+			loop i from: 1 to: 23{
+				
+				string path_grid;
+				if (i < 10){
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_O3_202105270' + i + '_202105270' + i + '.tif';
+				}
+				else{
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_O3_20210527' + i + '_20210527' + i + '.tif';
+				}
+				
+				matrix my_data <- grid_file(path_grid) as_matrix({624, 725});
+		        
+		        matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+					
+				loop j from: 0 to: 623{
+					loop k from: 0 to: 724{
+						
+						pollution_raster_float[j, k] <- float (my_data[j, k] get("grid_value"));
+						
+					}
+				}
+				write ('raster O3 ' + i + ' chargé');
+				write('max: ' + max(pollution_raster_float));
+				
+				O3_rasters << pollution_raster_float;
+			}
+		}
+		
+		if measure_PM10 {
+			
+			matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+			PM10_rasters << pollution_raster_float;
+			
+			loop i from: 1 to: 23{
+				
+				string path_grid;
+				if (i < 10){
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_PM10_202105270' + i + '_202105270' + i + '.tif';
+				}
+				else{
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_PM10_20210527' + i + '_20210527' + i + '.tif';
+				}
+				
+				matrix my_data <- grid_file(path_grid) as_matrix({624, 725});
+		        
+		        matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+					
+				loop j from: 0 to: 623{
+					loop k from: 0 to: 724{
+						
+						pollution_raster_float[j, k] <- float (my_data[j, k] get("grid_value"));
+						
+					}
+				}
+				write ('raster PM10 ' + i + ' chargé');
+				write('max: ' + max(pollution_raster_float));
+				
+				PM10_rasters << pollution_raster_float;
+			}
+		}
+		
+		if measure_PM25 {
+			
+			matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+			PM25_rasters << pollution_raster_float;
+			
+			loop i from: 1 to: 23{
+				
+				string path_grid;
+				if (i < 10){
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_PM25_202105270' + i + '_202105270' + i + '.tif';
+				}
+				else{
+					path_grid <- '../includes/Marseille/pollution_model/raster_dep13_PM25_20210527' + i + '_20210527' + i + '.tif';
+				}
+				
+				matrix my_data <- grid_file(path_grid) as_matrix({624, 725});
+		        
+		        matrix<float> pollution_raster_float <- 0.0 as_matrix({624, 725});
+					
+				loop j from: 0 to: 623{
+					loop k from: 0 to: 724{
+						
+						pollution_raster_float[j, k] <- float (my_data[j, k] get("grid_value"));
+						
+					}
+				}
+				write ('raster PM25 ' + i + ' chargé');
+				write('max: ' + max(pollution_raster_float));
+				
+				PM25_rasters << pollution_raster_float;
+			}
+		}
+		
+		
+		// generation des donnees de pollution pour chaque type d'agent
+		
+		if workers{
+	    	write("generating workers measures");
+	    		    	
+	    	create measure from: shape_file("../results/Marseille/worker/aller_worker/positions_aller_worker.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    		
+	    		int worker_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(worker_number);
+				if (worker_number <= 3000){do die;} 
+
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+				  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/worker/workers_measures_to_predict.csv" type:"csv" rewrite: false;
+	    	}
+	    	
+	    	ask measure{do die;}
+	    	
+	    	create measure from: shape_file("../results/Marseille/worker/retour_worker/positions_retour_worker.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    		
+	    		int worker_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(worker_number);
+				if (worker_number <= 3000){do die;} 
+
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/worker/workers_measures_to_predict.csv" type:"csv" rewrite: false;
+	    	}
+	    	ask measure{do die;}
+	    	
+	    		
+	    }	    	
+	    
+    	   		
+    	if students{
+	    	write("generating students measures");
+	    		    	
+	    	create measure from: shape_file("../results/Marseille/student/aller_student/positions_aller_student.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    		
+	    		int student_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(student_number);
+				if (student_number <= 1800){do die;} 
+				
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+				  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/student/students_measures_to_predict.csv" type:"csv" rewrite: false;
+	    	}
+	    	
+	    	ask measure{do die;}
+	    	
+	    	create measure from: shape_file("../results/Marseille/student/retour_student/positions_retour_student.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    			    		
+	    		int student_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(student_number);
+				if (student_number <= 1800){do die;} 
+	    		
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/student/students_measures_to_predict.csv" type:"csv" rewrite: false;
+	    	}
+	    	ask measure{do die;}
+	    		
+	    }
+	    
+	    if leisures{
+	    	write("generating leisures measures");
+	    		    	
+	    	create measure from: shape_file("../results/Marseille/leisure/aller_leisure/positions_aller_leisure.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    		
+	    		int leisure_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(leisure_number);
+				if (leisure_number <= 6200){do die;}
+
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+				  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/leisure/leisures_measures_to_predict.csv" type:"csv" rewrite: false;
+	    	}
+	    	
+	    	ask measure{do die;}
+	    	
+	    	create measure from: shape_file("../results/Marseille/leisure/retour_leisure/positions_retour_leisure.shp") with: [time_of_measure::(read("time")), name_of_agent::(read('agent'))]{
+	    		
+	    		int leisure_number <- int((((name_of_agent split_with '[')[1]) split_with ']')[0]);
+	    		write(leisure_number);
+				if (leisure_number <= 6200){do die;}
+	    		
+	    		ID <- compteur_ID;
+		    	compteur_ID <- compteur_ID + 1;
+		    			    					    
+			    geometry CRS_2154 <- location CRS_transform("EPSG:2154");
+			    
+			    list<string> coord <- string(CRS_2154) split_with ",";
+			    
+			    latitude <- float(coord[1]);
+			    longitude <- float((coord[0] split_with "{")[0]);
+							
+			    int cell_x <- int((longitude - min_x) / 25);
+			    int cell_y <- int((max_y - latitude) / 25);
+			    		    				    		
+			    // obtention de la concentration de polluant : interpolation linéaire entre les deux rasters horaires qui 'encadrent' l'heure de mesure
+			    
+				   	
+				int hour_before_measure <- time_of_measure.hour;
+				int hour_after_measure <- hour_before_measure + 1;
+					
+				int min_measure <- time_of_measure.minute;
+				int sec_measure <- time_of_measure.second;
+					
+				if measure_NO2 {	
+				NO2_concentration <- NO2_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (NO2_rasters[hour_after_measure][cell_x,cell_y] - NO2_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_O3 {	
+				O3_concentration <- O3_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (O3_rasters[hour_after_measure][cell_x,cell_y] - O3_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM10 {	
+				PM10_concentration <- PM10_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM10_rasters[hour_after_measure][cell_x,cell_y] - PM10_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				if measure_PM25 {	
+				PM25_concentration <- PM25_rasters[hour_before_measure][cell_x,cell_y] + 
+											  (PM25_rasters[hour_after_measure][cell_x,cell_y] - PM25_rasters[hour_before_measure][cell_x,cell_y]) * (min_measure * 60 + sec_measure) / 3600;
+											  
+				}
+				
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/leisure/leisures_measures_to_predict.csv" type:"csv" rewrite: false;
+	    	}
+	    	ask measure{do die;}
+	    		
+	    }		    	
+    }          
+}
+
+species measure {
+	int ID;
+	string name_of_agent;
+	float longitude min: min_x max: max_x - 0.1;
+	float latitude min: min_y max: max_y - 0.1;
+	date time_of_measure;
+	float NO2_concentration;
+	float O3_concentration;
+	float PM10_concentration;
+	float PM25_concentration;
+}
+
+experiment bike_traffic type: gui {
+
+}
\ No newline at end of file

8_Environment for measures to predict.gaml

+/**
+* Name: Deplacementdevelos
+* Based on the internal empty template. 
+* Author: nathancoisne
+* Tags: 
+*/
+
+
+model Deplacementdevelos
+
+/* 
+ * This model generates the environment close to the point of travel where the pollution level has to be predicted
+ * 
+ */
+
+global {
+	bool workers <- false;
+    bool students <- true;
+    bool leisures <- false;
+    	
+	file measures_workers_csv;
+    file measures_students_csv;
+    file measures_leisures_csv;
+    
+    file bound <- shape_file("../includes/Marseille/boundary_Marseille.shp");
+    geometry shape <- envelope(bound);
+    
+    file shape_file_buildings <- shape_file("../includes/Marseille/buildings_ign.shp");
+    
+    file shape_file_vegetation <- shape_file("../includes/Marseille/environment/ZONE_DE_VEGETATION.shp");
+    
+    file roads_ign <- shape_file("../includes/Marseille/environment/roads_ign.shp");
+    
+    file primary_roads <- shape_file("../includes/Marseille/environment/primary_roads.shp");
+
+
+    init{    	
+
+    	create building from: shape_file_buildings with: [height::float(read("HAUTEUR"))];
+    	
+    	create road from: roads_ign;
+    	list<road> rd_1_voie <- road where (each.NB_VOIES = 1);
+    	list<road> rd_2_voie <- road where (each.NB_VOIES = 2);
+    	list<road> rd_3_voie <- road where (each.NB_VOIES = 3);
+    	list<road> rd_4_voie <- road where (each.NB_VOIES = 4);
+    	list<road> rd_5_voie <- road where (each.NB_VOIES = 5);
+    	list<road> rd_6_voie <- road where (each.NB_VOIES = 6);
+    	
+    	list<road> list_rd_0_4_width <- road where (each.LARGEUR <= 4.0);
+    	list<road> list_rd_4_6_width <- road where (each.LARGEUR > 4.0 and each.LARGEUR <= 6.0);
+    	list<road> list_rd_6_8_width <- road where (each.LARGEUR > 6.0 and each.LARGEUR <= 8.0);
+    	list<road> list_rd_8_max_width <- road where (each.LARGEUR > 8.0);
+    	
+    	create primary_road from: primary_roads;
+    	
+    	create vegetation from: shape_file_vegetation;
+    	
+    	list<vegetation> list_bois <- vegetation where (each.NATURE = 'Bois');
+    	list<vegetation> list_foret <- vegetation where (each.NATURE = 'Forêt fermée de conifères' or each.NATURE = 'Forêt fermée de feuillus'
+    												     or each.NATURE = 'Forêt fermée mixte' or each.NATURE = 'Forêt ouverte');
+    	list<vegetation> list_haie <- vegetation where (each.NATURE = 'Haie');
+    	
+    	if(workers){
+    		measures_workers_csv <- csv_file("../results/Marseille/worker/workers_measures_to_predict.csv",true);
+    		
+	    	create measure from: measures_workers_csv{
+	    		write(self.ID);
+	    		location <- point(to_GAMA_CRS({self.longitude, self.latitude}, 'EPSG:2154'));
+	    		geometry disque <- circle(50, location);
+	    		
+	    		create environment {
+	    			ID <- myself.ID;
+	    			name_of_agent <- myself.name_of_agent;
+	    			longitude <- myself.longitude;
+	    			latitude <- myself.latitude;
+	    			
+	    			loop build over: building overlapping disque{
+						float surface_inter <- (build inter disque).area;
+						buildings_volume <- buildings_volume + build.height * surface_inter;
+					}
+					
+					loop rd over: rd_1_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_1 <- voie_1 + perimeter_inter;
+					}
+					
+					loop rd over: rd_2_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_2 <- voie_2 + perimeter_inter;
+					}
+					
+					loop rd over: rd_3_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_3 <- voie_3 + perimeter_inter;
+					}
+					
+					loop rd over: rd_4_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_4 <- voie_4 + perimeter_inter;
+					}
+					
+					loop rd over: rd_5_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_5 <- voie_5 + perimeter_inter;
+					}
+					
+					loop rd over: rd_6_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_6 <- voie_6 + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_0_4_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_0_4_width <- road_0_4_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_4_6_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_4_6_width <- road_4_6_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_6_8_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_6_8_width <- road_6_8_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_8_max_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_8_max_width <- road_8_max_width + perimeter_inter;
+					}
+					
+					loop veg over: list_bois overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						bois <- bois + area_inter;
+					}
+					
+					loop veg over: list_foret overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						foret <- foret + area_inter;
+					}
+					
+					loop veg over: list_haie overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						haie <- haie + area_inter;
+					}
+					
+					distance_to_main_road <- myself distance_to (primary_road closest_to myself);
+					
+	    		}
+	    	}
+	    	
+	    	ask environment {
+		    			save [ID, name_of_agent, longitude, latitude, buildings_volume, voie_1, voie_2, voie_3, voie_4, voie_5, voie_6, road_0_4_width, road_4_6_width, road_6_8_width, road_8_max_width, distance_to_main_road, bois, foret, haie]
+	    				to: "../results/Marseille/worker/workers_environment_for_prediction.csv" type:"csv" rewrite: false;   
+		    }   
+		}
+    	
+    	if(students){
+    		measures_students_csv <- csv_file("../results/Marseille/student/students_measures_to_predict.csv",true);
+    		
+	    	create measure from: measures_students_csv{
+	    		write(self.ID);
+	    		location <- point(to_GAMA_CRS({self.longitude, self.latitude}, 'EPSG:2154'));
+	    		geometry disque <- circle(50, location);
+	    		
+
+	    		create environment {
+	    			ID <- myself.ID;
+	    			name_of_agent <- myself.name_of_agent;
+	    			longitude <- myself.longitude;
+	    			latitude <- myself.latitude;
+	    			
+	    			loop build over: building overlapping disque{
+						float surface_inter <- (build inter disque).area;
+						buildings_volume <- buildings_volume + build.height * surface_inter;
+					}
+					
+					loop rd over: rd_1_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_1 <- voie_1 + perimeter_inter;
+					}
+					
+					loop rd over: rd_2_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_2 <- voie_2 + perimeter_inter;
+					}
+					
+					loop rd over: rd_3_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_3 <- voie_3 + perimeter_inter;
+					}
+					
+					loop rd over: rd_4_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_4 <- voie_4 + perimeter_inter;
+					}
+					
+					loop rd over: rd_5_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_5 <- voie_5 + perimeter_inter;
+					}
+					
+					loop rd over: rd_6_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_6 <- voie_6 + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_0_4_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_0_4_width <- road_0_4_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_4_6_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_4_6_width <- road_4_6_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_6_8_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_6_8_width <- road_6_8_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_8_max_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_8_max_width <- road_8_max_width + perimeter_inter;
+					}
+					
+					loop veg over: list_bois overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						bois <- bois + area_inter;
+					}
+					
+					loop veg over: list_foret overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						foret <- foret + area_inter;
+					}
+					
+					loop veg over: list_haie overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						haie <- haie + area_inter;
+					}
+					
+					distance_to_main_road <- myself distance_to (primary_road closest_to myself);
+					
+	    		}
+
+	    	}
+	    	
+	    	ask environment {
+		    			save [ID, name_of_agent, longitude, latitude, buildings_volume, voie_1, voie_2, voie_3, voie_4, voie_5, voie_6, road_0_4_width, road_4_6_width, road_6_8_width, road_8_max_width, distance_to_main_road, bois, foret, haie]
+	    				to: "../results/Marseille/student/students_environment_for_prediction.csv" type:"csv" rewrite: false;   
+		    }   
+		}
+		
+		if(leisures){
+    		measures_leisures_csv <- csv_file("../results/Marseille/leisure/leisures_measures_to_predict.csv",true);
+    		
+	    	create measure from: measures_leisures_csv{
+	    		write(self.ID);
+	    		location <- point(to_GAMA_CRS({self.longitude, self.latitude}, 'EPSG:2154'));
+	    		geometry disque <- circle(50, location);
+	    		
+	    		create environment {
+	    			ID <- myself.ID;
+	    			name_of_agent <- myself.name_of_agent;
+	    			longitude <- myself.longitude;
+	    			latitude <- myself.latitude;
+	    			
+	    			loop build over: building overlapping disque{
+						float surface_inter <- (build inter disque).area;
+						buildings_volume <- buildings_volume + build.height * surface_inter;
+					}
+					
+					loop rd over: rd_1_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_1 <- voie_1 + perimeter_inter;
+					}
+					
+					loop rd over: rd_2_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_2 <- voie_2 + perimeter_inter;
+					}
+					
+					loop rd over: rd_3_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_3 <- voie_3 + perimeter_inter;
+					}
+					
+					loop rd over: rd_4_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_4 <- voie_4 + perimeter_inter;
+					}
+					
+					loop rd over: rd_5_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_5 <- voie_5 + perimeter_inter;
+					}
+					
+					loop rd over: rd_6_voie overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						voie_6 <- voie_6 + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_0_4_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_0_4_width <- road_0_4_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_4_6_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_4_6_width <- road_4_6_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_6_8_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_6_8_width <- road_6_8_width + perimeter_inter;
+					}
+					
+					loop rd over: list_rd_8_max_width overlapping disque{
+						float perimeter_inter <- (rd inter disque).perimeter;
+						road_8_max_width <- road_8_max_width + perimeter_inter;
+					}
+					
+					loop veg over: list_bois overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						bois <- bois + area_inter;
+					}
+					
+					loop veg over: list_foret overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						foret <- foret + area_inter;
+					}
+					
+					loop veg over: list_haie overlapping disque{
+						float area_inter <- (veg inter disque).area;
+						haie <- haie + area_inter;
+					}
+					
+					distance_to_main_road <- myself distance_to (primary_road closest_to myself);
+					
+	    		}
+	    	}
+	    	
+	    	ask environment {
+		    			save [ID, name_of_agent, longitude, latitude, buildings_volume, voie_1, voie_2, voie_3, voie_4, voie_5, voie_6, road_0_4_width, road_4_6_width, road_6_8_width, road_8_max_width, distance_to_main_road, bois, foret, haie]
+	    				to: "../results/Marseille/leisure/leisures_environment_for_prediction.csv" type:"csv" rewrite: false;   
+		    }   
+		}
+	}
+
+}
+
+species measure { // seulement besoin de la localisation de la mesure, etant donne que l'environnement est suppose stationnaire
+	int ID;
+	string name_of_agent;
+	float longitude;
+	float latitude;
+}
+
+species environment{ // flotants correspondant à l'environnement dans un disque de 50m de rayon centré sur la mesure
+	int ID; // ID égal à celui de la mesure
+	string name_of_agent;
+	float longitude;
+	float latitude;
+	
+	float road_0_4_width; // longueur de route dans le disque ayant comme attribut : largeur <= 4 
+	float road_4_6_width; // longueur de route dans le disque ayant comme attribut : 4 < largeur <= 6 
+	float road_6_8_width; // longueur de route dans le disque ayant comme attribut : 6 < largeur <= 8 
+	float road_8_max_width; // longueur de route dans le disque ayant comme attribut : 8 < largeur
+	float voie_1;
+	float voie_2;
+	float voie_3;
+	float voie_4;
+	float voie_5;
+	float voie_6;
+	float buildings_volume;
+	float distance_to_main_road;
+	float bois; //surface de bois dans le disque
+	float foret;
+	float haie;
+}
+
+species road {
+	float LARGEUR;
+	int NB_VOIES;
+}
+
+species primary_road{
+}
+
+species building {
+	int id;
+    string type; 
+    string types_str;
+    list<string> types;
+    float height;        
+}
+
+species vegetation {
+	string NATURE; //Bois, Forêt fermée de conifères, Forêt fermée de feuillus, Forêt fermée mixte, Forêt ouverte, Haie, Lande ligneuse, Verger, Vigne (types BD TOPO)
+}
+
+
+
+experiment bike_traffic type: gui {
+}
\ No newline at end of file

9_Regression and prediction.gaml

+/**
+* Name: Deplacementdevelos
+* Based on the internal empty template. 
+* Author: nathancoisne
+* Tags: 
+*/
+
+
+model Deplacementdevelos
+
+/* 
+ * This model generates a regression between the pollution level given by "Synthetc measures from agents' travels" and the environment indicators generated in "Measures' environment"
+ * 
+ */
+
+global {
+		
+	bool workers <- false;
+    bool students <- true;
+    bool leisures <- false;
+	
+	file measures_workers_csv;
+    file measures_students_csv;
+    file measures_leisures_csv;
+
+	file environment_workers_csv;
+    file environment_students_csv;
+    file environment_leisures_csv;
+    
+    file measures_to_predict_workers_csv;
+    file measures_to_predict_students_csv;
+    file measures_to_predict_leisures_csv;
+
+	file environment_for_prediction_workers_csv;
+    file environment_for_prediction_students_csv;
+    file environment_for_prediction_leisures_csv;
+
+    init{ 
+
+       	//matrice de régression routes largeur: DATE, longueur de route 0-4, 4-6, 6-8, 8-max, NO2
+    	//matrice de régression routes voies: DATE, voies 1, 2, 3, 4, 5, 6, NO2
+    	//matrice de régression bâtiments: DATE, buildings_volume, NO2
+		//matrice de régression végétation: DATE, bois, foret, haie, NO2
+    	    	
+    	if(students){
+    		measures_students_csv <- csv_file("../results/Marseille/student/students_measures.csv",true); //mesures faites par 1800 agents student
+    		environment_students_csv <- csv_file("../results/Marseille/student/students_measures_environment.csv",true);
+    		
+    		create measure from: measures_students_csv;
+    		create environment from: environment_students_csv;
+    		    		
+    		list<float> pollution_NO2 ;
+    		list<float> ind_date;
+    		
+    		loop m over: measure {
+    			pollution_NO2 << m.NO2_concentration;
+    			string time_m <- (((m.time_of_measure split_with "'(")[1]) split_with ")'")[0];
+    			ind_date << float(date(time_m));
+    		}
+    	    		
+    		//création de la matrice de régression routes voies
+    		list<float> ind_voies_1;
+    		list<float> ind_voies_2;
+    		list<float> ind_voies_3;
+    		list<float> ind_voies_4;
+    		list<float> ind_voies_5;
+    		list<float> ind_voies_6;
+    		
+    		loop env over: environment {
+    			ind_voies_1 << env.voie_1;
+    			ind_voies_2 << env.voie_2;
+    			ind_voies_3 << env.voie_3;
+    			ind_voies_4 << env.voie_4;
+    			ind_voies_5 << env.voie_5;
+    			ind_voies_6 << env.voie_6;
+    		}
+    		
+    		matrix routes_voies_regression <- matrix(ind_date, ind_voies_1, ind_voies_2, ind_voies_3, ind_voies_4, ind_voies_5, ind_voies_6, pollution_NO2);
+    		      		
+    		regression NO2_voies_student <- build(routes_voies_regression);
+    		write("regression generee. debut de la prediction des trajets restants");
+    		
+    		// prediction des trajets non réalisés à partir de l'environnement de ces trajets
+    		
+    		measures_to_predict_students_csv <- csv_file("../results/Marseille/student/students_measures_to_predict.csv",true);
+    		environment_for_prediction_students_csv <- csv_file("../results/Marseille/student/students_environment_for_prediction.csv",true);
+    		
+    		create measure from: measures_to_predict_students_csv; // mesures à prédire pour les agents student restants
+    		create environment from: environment_for_prediction_students_csv;
+    		
+    		loop env over: environment {
+    			measure associated_measure <- measure[int(env)];
+    			string time_m <- (((associated_measure.time_of_measure split_with "'(")[1]) split_with ")'")[0];
+    			float predict_date <- float(date(time_m));
+    			measure[int(env)].predicted_NO2_concentration <- predict(NO2_voies_student, [predict_date, env.voie_1, env.voie_2, env.voie_3, env.voie_4, env.voie_5, env.voie_6]);
+    		}
+    		
+    		ask measure{
+			
+				save [ID, name_of_agent, longitude, latitude, time_of_measure, NO2_concentration, predicted_NO2_concentration, O3_concentration, PM10_concentration, PM25_concentration] 
+	    				to: "../results/Marseille/student/students_measures_predicted.csv" type:"csv" rewrite: false;
+	    				
+	    	} 		   	
+		}
+	}
+
+}
+
+species measure {
+	int ID;
+	string name_of_agent;
+	float longitude;
+	float latitude;
+	string time_of_measure;
+	float NO2_concentration;
+	float O3_concentration;
+	float PM10_concentration;
+	float PM25_concentration;
+	
+	float predicted_NO2_concentration;
+	float predicted_O3_concentration;
+	float predicted_PM10_concentration;
+	float predicted_PM25_concentration;
+}
+
+species environment{ // flotants correspondant à l'environnement dans un disque de 50m de rayon centré sur la mesure
+	int ID; // ID égal à celui de la mesure
+	string name_of_agent;
+	float longitude;
+	float latitude;
+	
+	float road_0_4_width; // longueur de route dans le disque ayant comme attribut : largeur <= 4 
+	float road_4_6_width; // longueur de route dans le disque ayant comme attribut : 4 < largeur <= 6 
+	float road_6_8_width; // longueur de route dans le disque ayant comme attribut : 6 < largeur <= 8 
+	float road_8_max_width; // longueur de route dans le disque ayant comme attribut : 8 < largeur
+	float voie_1;
+	float voie_2;
+	float voie_3;
+	float voie_4;
+	float voie_5;
+	float voie_6;
+	float buildings_volume;
+	float distance_to_main_road;
+	float bois; //surface de bois dans le disque
+	float foret;
+	float haie;
+}
+
+species position_travel {
+	
+}
+experiment regression_prediction type: gui {
+	output{
+    	display city_display type: opengl{
+    		species measure;
+    	}
+    	
+    }
+}
\ No newline at end of file

Agents' exposition.gaml

+/**
+* Name: Deplacementdevelos
+* Based on the internal empty template. 
+* Author: nathancoisne
+* Tags: 
+*/
+
+
+model Deplacementdevelos
+
+/* 
+ * This model calculates every population's exposition to each pollutant, thanks to measures generated in 'synthetic measures from agents' travels'
+ * 
+ */
+
+global {
+		
+    file measures_csv;
+    bool workers <- false; // une seule population en 'true' à la fois pour le stockage des mesures par type d'agent
+    bool students <- true;
+    bool leisures <- false;
+
+    init{    	
+    	
+    	if workers{
+    		write("loading workers measures");
+    		
+    		measures_csv <- csv_file("../results/Marseille/worker/workers_measures.csv",true);
+    		
+    		create measure from: measures_csv{
+	    		
+	    		if(self.ID = 100000 or self.ID = 200000 or self.ID = 300000 or self.ID = 400000 or self.ID = 500000) {write(self.ID);}	    	
+	    		
+	    	}
+	    	
+	    	write("calcul exposition of workers");
+	    	
+	    	loop i from: 0 to: 3600{
+	    		
+	    		write(i);
+	    		
+	    		list<measure> list_of_measures <- measure where (each.name_of_agent = 'worker[' + i + ']');
+	    		list<float> list_of_NO2;
+	    		list<float> list_of_O3;
+	    		list<float> list_of_PM10;
+	    		list<float> list_of_PM25;
+	    		
+	    		loop m over: list_of_measures{
+	    			if (m.NO2_concentration > 0) {list_of_NO2 << m.NO2_concentration;}
+	    			if (m.O3_concentration > 0) {list_of_O3 << m.O3_concentration;}
+	    			if (m.PM10_concentration > 0) {list_of_PM10 << m.PM10_concentration;}
+	    			if (m.PM25_concentration > 0) {list_of_PM25 << m.PM25_concentration;}
+	    		}
+	    		
+	    		create worker{
+	    			name_of_agent <- 'worker[' + i + ']';
+	    			
+	    			NO2_mean_exposition <- mean (list_of_NO2);
+	    			O3_mean_exposition <- mean (list_of_O3);
+	    			PM10_mean_exposition <- mean (list_of_PM10);
+	    			PM25_mean_exposition <- mean (list_of_PM25);
+	    			
+	    			NO2_max <- max (list_of_NO2);
+	    			O3_max <- max (list_of_O3);
+	    			PM10_max <- max (list_of_PM10);
+	    			PM25_max <- max (list_of_PM25);
+	    			
+	    		}	
+	    	}
+	    	ask worker {
+		    			save [name_of_agent, NO2_mean_exposition, NO2_max, O3_mean_exposition, O3_max, PM10_mean_exposition, PM10_max, PM25_mean_exposition, PM25_max] 
+	    				to: "../results/Marseille/worker/workers_exposition.csv" type:"csv" rewrite: false;
+		    	}
+    	}
+    	
+    	if students{
+    		write("loading students measures");
+    		
+    		measures_csv <- csv_file("../results/Marseille/student/students_measures.csv",true);
+    		
+    		create measure from: measures_csv{
+	    		
+	    		if(self.ID = 100000 or self.ID = 200000 or self.ID = 300000 or self.ID = 400000 or self.ID = 500000) {write(self.ID);}	    	
+	    		
+	    	}
+	    	
+	    	write("calcul exposition of students");
+	    	
+	    	loop i from: 0 to: 2236{
+	    		
+	    		write(i);
+	    		
+	    		list<measure> list_of_measures <- measure where (each.name_of_agent = 'student[' + i + ']');
+	    		list<float> list_of_NO2;
+	    		list<float> list_of_O3;
+	    		list<float> list_of_PM10;
+	    		list<float> list_of_PM25;
+	    		
+	    		loop m over: list_of_measures{
+	    			if (m.NO2_concentration > 0) {list_of_NO2 << m.NO2_concentration;}
+	    			if (m.O3_concentration > 0) {list_of_O3 << m.O3_concentration;}
+	    			if (m.PM10_concentration > 0) {list_of_PM10 << m.PM10_concentration;}
+	    			if (m.PM25_concentration > 0) {list_of_PM25 << m.PM25_concentration;}
+	    		}
+	    		
+	    		create student{
+	    			name_of_agent <- 'student[' + i + ']';
+	    			
+	    			NO2_mean_exposition <- mean (list_of_NO2);
+	    			O3_mean_exposition <- mean (list_of_O3);
+	    			PM10_mean_exposition <- mean (list_of_PM10);
+	    			PM25_mean_exposition <- mean (list_of_PM25);
+	    			
+	    			NO2_max <- max (list_of_NO2);
+	    			O3_max <- max (list_of_O3);
+	    			PM10_max <- max (list_of_PM10);
+	    			PM25_max <- max (list_of_PM25);
+	    			
+	    		}	
+	    	}
+	    	ask student {
+		    			save [name_of_agent, NO2_mean_exposition, NO2_max, O3_mean_exposition, O3_max, PM10_mean_exposition, PM10_max, PM25_mean_exposition, PM25_max] 
+	    				to: "../results/Marseille/student/students_exposition.csv" type:"csv" rewrite: false;
+		    	}
+    	}
+    	
+    	if leisures{
+    		write("loading leisures measures");
+    		
+    		measures_csv <- csv_file("../results/Marseille/worker/leisures_measures.csv",true);
+    		
+    		create measure from: measures_csv{
+	    		
+	    		if(self.ID = 100000 or self.ID = 200000 or self.ID = 300000 or self.ID = 400000 or self.ID = 500000) {write(self.ID);}	    	
+	    		
+	    	}
+	    	
+	    	write("calcul exposition of leisures");
+	    	
+	    	loop i from: 0 to: 7527{
+	    		
+	    		write(i);
+	    		
+	    		list<measure> list_of_measures <- measure where (each.name_of_agent = 'leisure[' + i + ']');
+	    		list<float> list_of_NO2;
+	    		list<float> list_of_O3;
+	    		list<float> list_of_PM10;
+	    		list<float> list_of_PM25;
+	    		
+	    		loop m over: list_of_measures{
+	    			if (m.NO2_concentration > 0) {list_of_NO2 << m.NO2_concentration;}
+	    			if (m.O3_concentration > 0) {list_of_O3 << m.O3_concentration;}
+	    			if (m.PM10_concentration > 0) {list_of_PM10 << m.PM10_concentration;}
+	    			if (m.PM25_concentration > 0) {list_of_PM25 << m.PM25_concentration;}
+	    		}
+	    		
+	    		create leisure{
+	    			name_of_agent <- 'leisure[' + i + ']';
+	    			
+	    			NO2_mean_exposition <- mean (list_of_NO2);
+	    			O3_mean_exposition <- mean (list_of_O3);
+	    			PM10_mean_exposition <- mean (list_of_PM10);
+	    			PM25_mean_exposition <- mean (list_of_PM25);
+	    			
+	    			NO2_max <- max (list_of_NO2);
+	    			O3_max <- max (list_of_O3);
+	    			PM10_max <- max (list_of_PM10);
+	    			PM25_max <- max (list_of_PM25);
+	    			
+	    		}	
+	    	}
+	    	ask leisure {
+		    			save [name_of_agent, NO2_mean_exposition, NO2_max, O3_mean_exposition, O3_max, PM10_mean_exposition, PM10_max, PM25_mean_exposition, PM25_max] 
+	    				to: "../results/Marseille/leisure/leisures_exposition.csv" type:"csv" rewrite: false;
+		    	}
+    	}
+	}
+
+}
+
+species measure { // seulement besoin de la localisation de la mesure, etant donne que l'environnement est suppose stationnaire
+	string name_of_agent;
+	int ID;
+	float NO2_concentration;
+	float O3_concentration;
+	float PM10_concentration;
+	float PM25_concentration;
+}
+
+species people {
+	string name_of_agent;
+	
+	float NO2_mean_exposition;
+	float O3_mean_exposition;
+	float PM10_mean_exposition;
+	float PM25_mean_exposition; 
+	
+	float NO2_max;
+	float O3_max;
+	float PM10_max;
+	float PM25_max;
+}
+
+species worker parent: people{
+
+}
+
+species student parent: people{
+	
+}
+
+species leisure parent: people{
+	
+}
+
+experiment bike_traffic type: gui {
+	output{
+    	display exposition {
+    		chart "Exposition moyenne des étudiants" type: histogram background: #lightgrey{
+    			data "0 - 40 um/m3" value: student count (each.NO2_mean_exposition < 40) ;
+    			
+    			data "40 - 80 um/m3" value: student count (each.NO2_mean_exposition >= 40 and each.NO2_mean_exposition < 80);
+    			
+    			data "80 - 120 um/m3" value: student count (each.NO2_mean_exposition >= 80 and each.NO2_mean_exposition < 120);
+
+				data "120 - 160 um/m3" value: student count (each.NO2_mean_exposition >= 120 and each.NO2_mean_exposition < 160);
+				
+				data "160 - 200 um/m3" value: student count (each.NO2_mean_exposition >= 160 and each.NO2_mean_exposition < 200);
+    		}
+
+    	}
+	}
+
+}
\ No newline at end of file

Digital Elevation Model for roads.gaml

+/***
+* Part of the SWITCH Project
+* Author: Patrick Taillandier
+* Tags: gis, OSM data
+***/
+
+model switch_utilities_gis
+/*
+ * Bases de données requises : 
+ * https://download.geofabrik.de/europe/france.html
+ * https://geoservices.ign.fr/documentation/diffusion/telechargement-donnees-libres.html#bd-topo
+ * 
+ */
+ 
+global {
+	
+	string dataset_path <- "../includes/";
+		
+	file shape_file_roads <- shape_file("../includes/SHP Marseille/roads.shp");
+	
+	file elevation_data <- grid_file('../includes/DEM_Marseille.tif');
+							
+	geometry shape <- envelope(elevation_data);
+	
+	graph the_graph;
+            
+    int nb_for_road_shapefile_split <- 20000;
+    
+	
+	init {
+		float max_value <- elevation_cell max_of (each.grid_value);
+		float min_value <- elevation_cell min_of (each.grid_value);
+		ask elevation_cell {
+			normalized_elevation <- (grid_value) / (max_value);
+			if grid_value < -1{
+				color <- #black;
+			}
+			else{
+				color <- rgb(int(255 * normalized_elevation), int(255 * (normalized_elevation)), int(255 * (normalized_elevation)));
+				
+			}
+		}
+				
+		write "Start the pre-processing process";
+
+    	create road from: shape_file_roads with: [id::int(get("id")), fclass::get("fclass"), speed_coeff::float(get("s_coeff")),maxspeed::int(get("maxspeed"))]{
+    		point first <- first(shape.points);
+			point last <- last(shape.points);
+    		if (not (first overlaps world)) or (not (last overlaps world)) {
+				do die;
+			}
+			
+			if length(shape.points) > 1{
+				float altitude_first <- (elevation_cell where (first overlaps each))[0].grid_value; //(distance_to (first, each.location))).grid_value;
+	    		float altitude_last <- (elevation_cell where (last overlaps each))[0].grid_value; //(distance_to (last, each.location))).grid_value;	
+	    		
+	    		if (altitude_last > -3 and altitude_first > -3 and shape.perimeter > 0){
+    				slope <- sin((altitude_last - altitude_first) / shape.perimeter) * 180 / #pi;
+    			}
+	    	}
+    	}
+    	
+		write "Number of roads created : "+ length(road);    	
+		
+		write("Saving roads");   
+		 	
+		save road to:"../includes/SHP Marseille/roads_slope.shp" type: shp attributes: ["id"::id,"fclass"::fclass, "s_coeff"::speed_coeff, "maxspeed"::maxspeed, "slope"::slope];
+			
+		
+		write("over");
+
+	}
+	
+}
+
+grid elevation_cell file: elevation_data {
+	float normalized_elevation min:0.0 max: 1.0;
+}
+
+
+species road  {
+	int id;
+	int maxspeed;
+	string fclass;
+	float speed_coeff;
+	float slope;
+		
+	rgb color <- #red;
+	
+    aspect base {
+    draw shape color: color width: 2;
+    }
+}
+
+
+experiment generateGISdata type: gui {
+	output {
+		display map type: opengl draw_env: false{
+			species elevation_cell;
+			species road;
+		}
+	}
+}
\ No newline at end of file

Synthetic population viewer.gaml

+/**
+* Name: Deplacementdevelos
+* Based on the internal empty template. 
+* Author: nathancoisne
+* Tags: 
+*/
+
+
+model Deplacementdevelos
+
+/* Modélisation du déplacement d'une flotte de cyclistes équipés de capteurs de pollution
+ * Sortie : une carte des zones mesurées est générée toutes les heures
+ * Sortie : un compte rendu du trajet de chaque cycliste, avec des points de rendez-vous 
+ * 
+ */
+
+global {
+	
+	string city <- "Marseille";
+	bool test_population <- false;
+	
+	file bound <- shape_file("../includes/" + city + "/boundary_" + city + ".shp");
+	
+	file shape_file_buildings;
+                            
+    file shape_file_roads;
+	
+	
+	bool workers <- false;
+    bool students <- true;
+    bool leisures <- false;
+    bool deliveries <- false;
+    
+    bool home_location <- true;
+    
+    file worker_csv;
+    
+    file student_csv;
+    
+    file leisure_csv;
+    
+    file delivery_midi_csv;
+    
+    file delivery_soir_csv;
+        
+    geometry shape <- envelope(bound);
+       	        	
+    graph the_graph;
+            
+    list<string> study_places <- ["university", "college", "school", "library"];
+ 
+    init{    	
+    	if city = "Marseille"{
+    		shape_file_buildings <- shape_file("../includes/Marseille/SHP Marseille/buildings_v2.shp"); // building contenant les data osm, ign, pois
+    		shape_file_roads <- shape_file("../includes/Marseille/SHP Marseille/roads.shp");
+    		
+    		if (test_population){
+	    		write("Marseille : population test");
+	    		worker_csv <- csv_file( "../results/Marseille/synthetic_population_test/worker.csv",true);
+	    
+			    student_csv <- csv_file( "../results/Marseille/synthetic_population_test/student.csv",true);
+			    
+			    leisure_csv <- csv_file( "../results/Marseille/synthetic_population_test/leisure.csv",true);
+			    
+			    delivery_midi_csv <- csv_file( "../results/Marseille/synthetic_population_test/delivery_midi.csv",true);
+			    
+			    delivery_soir_csv <- csv_file( "../results/Marseille/synthetic_population_test/delivery_soir.csv",true);
+	    	}
+    		else{
+	    		write("Marseille : population reelle");
+	    		worker_csv <- csv_file( "../results/Marseille/synthetic_population/worker.csv",true);
+	    
+			    student_csv <- csv_file( "../results/Marseille/synthetic_population/student.csv",true);
+			    
+			    leisure_csv <- csv_file( "../results/Marseille/synthetic_population/leisure.csv",true);
+			    
+			    delivery_midi_csv <- csv_file( "../results/Marseille/synthetic_population/delivery_midi.csv",true);
+			    
+			    delivery_soir_csv <- csv_file( "../results/Marseille/synthetic_population/delivery_soir.csv",true);
+	    	}
+    	}
+    	
+    	if city = "Toulouse"{
+    		shape_file_buildings <- shape_file("../includes/Toulouse/buildings_with_pois.shp"); // building contenant les data osm, ign, pois
+    		shape_file_roads <- shape_file("../includes/Toulouse/roads.shp");
+    		
+    		if (test_population){
+	    		write("Toulouse : population test");
+	    		worker_csv <- csv_file( "../results/Toulouse/synthetic_population_test/worker.csv",true);
+	    
+			    student_csv <- csv_file( "../results/Toulouse/synthetic_population_test/student.csv",true);
+			    
+			    leisure_csv <- csv_file( "../results/Toulouse/synthetic_population_test/leisure.csv",true);
+			    
+			    delivery_midi_csv <- csv_file( "../results/Toulouse/synthetic_population_test/delivery_midi.csv",true);
+			    
+			    delivery_soir_csv <- csv_file( "../results/Toulouse/synthetic_population_test/delivery_soir.csv",true);
+	    	}
+    		else{
+	    		write("Toulouse : population reelle");
+	    		worker_csv <- csv_file( "../results/Toulouse/synthetic_population/worker.csv",true);
+	    
+			    student_csv <- csv_file( "../results/Toulouse/synthetic_population/student.csv",true);
+			    
+			    leisure_csv <- csv_file( "../results/Toulouse/synthetic_population/leisure.csv",true);
+			    
+			    delivery_midi_csv <- csv_file( "../results/Toulouse/synthetic_population/delivery_midi.csv",true);
+			    
+			    delivery_soir_csv <- csv_file( "../results/Toulouse/synthetic_population/delivery_soir.csv",true);
+	    	}
+    	}
+    	
+
+    	create building from: shape_file_buildings {
+    		types <- types_str split_with ",";
+
+    	}
+    	
+    	//list<building> university_buildings <- building where (not empty(study_places inter each.types));
+       	    	
+    	create road from: shape_file_roads with: [id::int(get("id")), fclass::read ("fclass"), speed_coeff::float(get("s_coeff")), maxspeed::int(get("maxspeed"))];
+    	
+    	the_graph <- as_edge_graph(road);
+		the_graph <- the_graph;    	
+    	
+    	map<int,building> building_map <- building as_map (each.id::each);
+
+		if workers{
+	    	write("instanciate workers");
+	    	create worker from: worker_csv with:[id_living_place::int(get("living_place.id")), id_destination_place::int(get("destination_place.id"))]{
+	    		
+	    		if home_location{
+	    											   	
+		    		living_place <- building_map[id_living_place];																			 
+		    		location <- any_location_in (living_place);
+	    		}
+	    		else {
+	    			destination_place <- building_map[id_destination_place];
+	    			location <- any_location_in (destination_place);
+	    		}
+
+	    	}
+    	}
+    	
+    	if students{
+	    	write("instanciate students");
+	    	create student from: student_csv with:[id_living_place::int(get("living_place.id")), id_destination_place::int(get("destination_place.id"))]{
+	    																		   	
+	    		if home_location{
+	    											   	
+		    		living_place <- building_map[id_living_place];																			 
+		    		location <- any_location_in (living_place);
+	    		}
+	    		else {
+	    			destination_place <- building_map[id_destination_place];
+	    			location <- any_location_in (destination_place);
+	    		}
+	  	    }
+	    }
+
+		if leisures{
+	    	write("instanciate leisures");
+	    	create leisure from: leisure_csv with:[id_living_place::int(get("living_place.id")), id_destination_place::int(get("destination_place.id"))]{
+	    																						   	
+	    		if home_location{
+	    											   	
+		    		living_place <- building_map[id_living_place];																			 
+		    		location <- any_location_in (living_place);
+	    		}
+	    		else {
+	    			destination_place <- building_map[id_destination_place];
+	    			location <- any_location_in (destination_place);
+	    		}
+	    	}
+	    }
+
+    
+    	if deliveries {
+	    	write("instanciate deliveries");
+			create delivery from: delivery_midi_csv with: [id_living_place::int(get("living_place.id"))]{
+			   	
+	    		living_place <- building_map[id_living_place];																				 
+	    		location <- any_location_in (living_place);
+	    		
+
+			}   
+			
+	    	create delivery from: delivery_soir_csv with: [id_living_place::int(get("living_place.id"))]{
+	   					   				    			   	
+	    		living_place <- building_map[id_living_place];																				 
+	    		location <- any_location_in (living_place);
+
+			}   
+		}
+		
+		/*
+		float min_dist <- university_buildings min_of (each distance_to student[0]);
+		float max_dist <- university_buildings max_of (each distance_to student[0]);
+		
+		ask building {
+			if not (self in university_buildings){
+				do die;
+			}
+			float coeff_dist <- (self distance_to student[0] - min_dist) / (max_dist - min_dist);
+			
+			color <- rgb(int(255 * coeff_dist), int(255 * (1 - coeff_dist)), 0);
+		}
+		* 
+		*/
+    }          
+}
+
+species building {
+	int id;
+    string type; 
+    string types_str;
+    list<string> types;
+    float height;
+    rgb color <- #gray;
+        
+    aspect base {
+    draw shape color: color ;
+    }
+}
+
+species road  {
+	int id;
+	
+	string fclass;
+	
+	int maxspeed;
+	
+	float speed_coeff;
+	
+	rgb color <- #orange; 
+		
+    aspect base {
+    draw shape color: color width: 1;
+    }
+}
+
+species people skills: [moving] {
+	rgb color;
+	
+	int id_living_place;
+	int id_destination_place;
+	
+	building living_place;
+	building destination_place;
+	
+	aspect base{
+		draw circle(60) color: color border: #black;
+	}
+}
+
+species worker parent: people {
+	rgb color <- #black;
+}
+
+species student parent: people{
+	rgb color <- #green;
+}
+
+species leisure parent: people{
+	rgb color <- #blue;
+}
+
+species delivery skills: [moving]{
+	rgb color <- #red;
+	
+	int id_living_place;
+	building living_place <- nil;
+    
+	aspect base{
+		draw circle(10) color: color border: #black;
+	}
+}
+
+
+
+experiment bike_traffic type: gui {
+       
+    output{
+    	display city_display type: opengl{
+    		species building aspect: base;
+    		//species road aspect: base;
+    		species worker aspect: base;
+    		species student aspect: base;
+    		species leisure aspect: base;
+    		species delivery aspect: base;	    	
+    	}
+    	
+    	display caract_road {
+    		chart "Distribution de la distance de trajet des travailleurs" type: histogram background: #lightgrey size: {0.5, 0.5} position: {0, 0}{
+				data "0 - 50 m" value: road count (each.shape.perimeter < 50 #m) color: #blue;
+				
+				data "50 - 100" value: road count (each.shape.perimeter > 50 #m and each.shape.perimeter < 100 #m) color: #blue;
+				
+				data "100 - 150" value: road count (each.shape.perimeter > 100 #m and each.shape.perimeter < 150 #m) color: #blue;
+				
+				data "150 - 200" value: road count (each.shape.perimeter > 150 #m and each.shape.perimeter < 200 #m) color: #blue;
+				
+			}
+			
+			chart "Distance moyenne des routes" type: histogram background: #lightgrey size: {0.5, 0.5} position: {0.5, 0.5} {
+    			list<float> size;
+    		
+	    		loop rd over: road{
+	    			size << rd.shape.perimeter;
+	    		}
+	    		data "Travail" value: sum(size) / length(road) color: #blue;
+   
+		
+    		}
+    	}
+	}
+	
+	
+	
+	
+	
+	
+	
+}
\ No newline at end of file

Travels study.gaml

+/**
+* Name: Deplacementdevelos
+* Based on the internal empty template. 
+* Author: nathancoisne
+* Tags: 
+*/
+
+
+model Deplacementdevelos
+
+/* 
+ * This model generates travels' characteristics (duration and length for comparison with real world data
+ * 
+ */
+
+global {
+		
+	file bound <- shape_file("../includes/Marseille/boundary_Marseille.shp");
+		  	
+	bool workers <- true;
+    bool students <- false; // une seule population à la fois 
+    bool leisures <- false;           	
+
+    init{ 
+    	
+    	if workers{
+	    	write("workers travels");
+	    	loop i from: 0 to: 3600 {
+	    		write(i);
+	    		
+	    		list<position> list_of_positions_aller <- [];
+	    		create travel{
+		    		create position from: shape_file("../results/Marseille/worker/aller_worker/positions/aller_worker[" + i + "].shp") with: [time_of_measure::(read("time"))]{
+	
+		    			list_of_positions_aller << self;
+		    		}
+		    		float milliseconds <- first(list_of_positions_aller).time_of_measure milliseconds_between last(list_of_positions_aller).time_of_measure;
+		    		
+		    		duration <- int(milliseconds / 1000);
+		  			
+		  			distance_vol_oiseau <- int(distance_to(first(list_of_positions_aller), last(list_of_positions_aller)));
+		  			
+		  			float dist;		
+			    	create road from: shape_file("../results/Marseille/worker/aller_worker/trajets/aller_worker[" + i + "].shp"){
+			    		dist <- dist + self.shape.perimeter;
+			    	}
+			    	
+			    	distance <- int(dist);
+		    	}
+		    	
+		    	list<position> list_of_positions_retour <- [];
+	    		create travel{
+		    		create position from: shape_file("../results/Marseille/worker/retour_worker/positions/retour_worker[" + i + "].shp") with: [time_of_measure::(read("time"))]{
+	
+		    			list_of_positions_retour << self;
+		    		}
+		    		float milliseconds <- first(list_of_positions_retour).time_of_measure milliseconds_between last(list_of_positions_retour).time_of_measure;
+		    		
+		    		duration <- int(milliseconds / 1000);
+		  			
+		  			distance_vol_oiseau <- int(distance_to(first(list_of_positions_retour), last(list_of_positions_retour)));
+		  			
+		  			float dist;		
+			    	create road from: shape_file("../results/Marseille/worker/retour_worker/trajets/retour_worker[" + i + "].shp"){
+			    		dist <- dist + self.shape.perimeter;
+			    	}
+			    	
+			    	distance <- int(dist);
+		    	}	
+	    	}
+	    	ask travel{
+				save [distance_vol_oiseau, distance, duration] to: "../results/Marseille/worker/workers_travel_time.csv" type:"csv" rewrite: false;
+			}
+    	}
+    	
+    	if students{
+	    	write("students travels");
+	    	loop i from: 0 to: 2236 {
+	    		write(i);
+	    		
+	    		list<position> list_of_positions_aller <- [];
+	    		create travel{
+		    		create position from: shape_file("../results/Marseille/student/aller_student/positions/aller_student[" + i + "].shp") with: [time_of_measure::(read("time"))]{
+	
+		    			list_of_positions_aller << self;
+		    		}
+		    		float milliseconds <- first(list_of_positions_aller).time_of_measure milliseconds_between last(list_of_positions_aller).time_of_measure;
+		    		
+		    		duration <- int(milliseconds / 1000);
+		  			
+		  			distance_vol_oiseau <- int(distance_to(first(list_of_positions_aller), last(list_of_positions_aller)));
+		  			
+		  			float dist;		
+			    	create road from: shape_file("../results/Marseille/student/aller_student/trajets/aller_student[" + i + "].shp"){
+			    		dist <- dist + self.shape.perimeter;
+			    	}
+			    	
+			    	distance <- int(dist);
+		    	}
+		    	
+		    	list<position> list_of_positions_retour <- [];
+	    		create travel{
+		    		create position from: shape_file("../results/Marseille/student/retour_student/positions/retour_student[" + i + "].shp") with: [time_of_measure::(read("time"))]{
+	
+		    			list_of_positions_retour << self;
+		    		}
+		    		float milliseconds <- first(list_of_positions_retour).time_of_measure milliseconds_between last(list_of_positions_retour).time_of_measure;
+		    		
+		    		duration <- int(milliseconds / 1000);
+		  			
+		  			distance_vol_oiseau <- int(distance_to(first(list_of_positions_retour), last(list_of_positions_retour)));
+		  			
+		  			float dist;		
+			    	create road from: shape_file("../results/Marseille/student/aller_student/trajets/aller_student[" + i + "].shp"){
+			    		dist <- dist + self.shape.perimeter;
+			    	}
+			    	
+			    	distance <- int(dist);
+		    	}	
+	    	}
+	    	ask travel{
+				save [distance_vol_oiseau, distance, duration] to: "../results/Marseille/student/students_travel_time.csv" type:"csv" rewrite: false;
+			}
+    	}
+    }
+}
+
+species travel {
+	int duration; //temps de trajet en secondes
+	
+	int distance_vol_oiseau; //distance du trajet à vol d'oiseau en metres
+	
+	int distance;
+}
+
+species position {
+	date time_of_measure;
+}
+
+species road  {
+	string fclass;
+	rgb color <- #blue;
+}
+
+
+experiment bike_traffic type: gui {
+	output{
+
+	}	
+}
\ No newline at end of file