Simulace.info - User contributions [en]

Finding strategies comparison

2023-01-23T18:49:29Z

Kadt02: /* Second method - bouncing */

This page presents the multi-agent simulation of comparison of strategies for finding a lost person in the forest made by Tomas Kadane.

=Problem definition=
This problem came to my mind because I have once lost my dog in a forest. I was thinking what would be the best strategy to find the dog. However it is pretty complex problem to simulate forest and all types of strategies, so this work is simplified by multiple factors (against the reality and work assignment). I'm comparing two strategies with variable number of searchers in square pane with fixed boundaries.

=Method=
For the simulation I have used NetLogo 6.3.0. I'm implementing two basic searching strategies. First one is based on random walk and the second is based on the walking to the edge of the forest and then bouncing in the opposite direction.

Searched person is represented by single red colored patch (representing area where the searched person can be seen). And the searchers are so called turtles beginning at random position. When any of the turtle enters the red area, the simulation stops and number of ticks is recorded. Both methods have been done 30 times for 1, 2 and 3 searchers.

==First method - random walk==
In the fist method I implemented random walk. Each step (represented by one tick) one step ahead is made and then the heading is changed by random angle using this formula:
set heading (heading + (angle / 2) - (random angle))

If the searchers hits the border of the forest, he will turn "backwards" as described in this code:
to check-borders
if (xcor < min-pxcor) [set heading (heading + 180)]
if (xcor > max-pxcor) [set heading (heading + 180)]
if (ycor < min-pycor) [set heading (heading + 180)]
if (ycor > max-pycor) [set heading (heading + 180)]
end

[[File:randomwalk_UI.png]]

After measuring 30 times for each angle and 1, 2 and 3 searchers I created table of average steps (ticks) needed to find the searched person. Not agregated data can be found in [[File:Results random walk.xlsx]]

{| class="wikitable"
|+ align="top" | '''Averages for random walk depending on the angle'''
!| Searchers || 45 angle || 90 angle || 180 angle || 360 angle
|-
| 1 || 1033 || 583 || 633 || 1301
|-
| 2 || 398 || 304 || 403 || 658
|-
| 3 || 290 || 219 || 130 || 371
|}

To see how steps needed vary I used standart deviation.

{| class="wikitable"
|+ align="top" | '''Standart deviations for random walk depending on the angle'''
!| Searchers || 45 angle || 90 angle || 180 angle || 360 angle
|-
| 1 || 885 || 543 || 662 || 1793
|-
| 2 || 377 || 236 || 356 || 869
|-
| 3 || 366 || 237 || 128 || 472
|}

==Second method - bouncing==
In the second method searchers and searchers are spawned on the random position. Setup is same as in the first method. However the moving method differs. Searchers are moving in direct way and when they hit border they bounce in oposite direction as described in this code:
to check-bounce
;; bounce off left and right walls
if abs pxcor = max-pxcor [
set heading (- heading)
]
;; bounce off top and bottom walls
if abs pycor = max-pycor [
set heading (180 - heading)
]
end

[[File:bounce_UI.png]]

The method of recording steps needed to find the lost person was same as in the first method, however the angle is not variable now. To see how the data vary I have used standart deviation again. Non agregated data can be seen here [[File:Results bounce.xlsx]]

{| class="wikitable"
|+ align="top" | '''Averages and standart deviations for bouncing method'''
!| Searchers || Average || St. dev.
|-
| 1 || 566 || 624
|-
| 2 || 192 || 161
|-
| 3 || 180 || 141
|}

=Results=
To evaluate random walk method. We can see that if we are searching alone, the 90 angle seems the most efective, however 180 angle seems not bad at all too. If we are searching as two people - 90 angle is the most suitable. When there are three searchers the 180 angle seems the best.

When we want to compare our two methods, For one searcher bounce method is better even then the most effective angle in random walk method. For two searchers it is the same, bounce method is more effective again, even then the 90 angle method. For three searchers is the random walk with the 180 angle most effective.

However if we want to evaluate the standart deviation, basicly it is always better with the bounce method. So it seems like safer choice if we don't want to risk really long journey to find the dog.

=Conclussion=
Although we can see in the tables, that there are differences in both strategies we mustn't forget, that we are using only 30 observation for all the variants. The number should be more significant to say that certainly. However we can see significant difference in standart deviation and thus we can see that using random element in our searching strategy can pay off but also can be pretty bad.

=Code=
==Random walk method==
globals [
found
]

to setup
clear-all
reset-ticks
set found false

setup-searchers
setup-lost-area
end

to setup-searchers
create-turtles searchers [
pen-up
]
ask turtles [
setxy random-xcor random-ycor
]
end

to setup-lost-area
ask patch random-xcor random-ycor [
set pcolor red
]

end

to check-borders
if (xcor < min-pxcor) [set heading (heading + 180)]
if (xcor > max-pxcor) [set heading (heading + 180)]
if (ycor < min-pycor) [set heading (heading + 180)]
if (ycor > max-pycor) [set heading (heading + 180)]
end

to go-random
if found = true [
stop
]
tick
ask turtles [
if pcolor = red [
set found true
]
check-borders
set heading (heading + (angle / 2) - (random angle))
forward 1
]
end

[[File:Sp random walk.nlogo]]

==Bounce method==
globals [
found
]

to setup
clear-all
reset-ticks
set found false

setup-searchers
setup-lost-area
end

to setup-searchers
create-turtles searchers [
pen-up
]
ask turtles [
setxy random-xcor random-ycor
]
end

to setup-lost-area
ask patch random-xcor random-ycor [
set pcolor red
]

end

to check-bounce
;; bounce off left and right walls
if abs pxcor = max-pxcor [
set heading (- heading)
]
;; bounce off top and bottom walls
if abs pycor = max-pycor [
set heading (180 - heading)
]
end

to go
if found = true [
stop
]
tick
ask turtles [
if pcolor = red [
set found true
]
check-bounce
forward 1
]
end

[[File:Sp bounce.nlogo]]

Finding strategies comparison

2023-01-23T18:48:54Z

Kadt02:

This page presents the multi-agent simulation of comparison of strategies for finding a lost person in the forest made by Tomas Kadane.

=Problem definition=
This problem came to my mind because I have once lost my dog in a forest. I was thinking what would be the best strategy to find the dog. However it is pretty complex problem to simulate forest and all types of strategies, so this work is simplified by multiple factors (against the reality and work assignment). I'm comparing two strategies with variable number of searchers in square pane with fixed boundaries.

=Method=
For the simulation I have used NetLogo 6.3.0. I'm implementing two basic searching strategies. First one is based on random walk and the second is based on the walking to the edge of the forest and then bouncing in the opposite direction.

Searched person is represented by single red colored patch (representing area where the searched person can be seen). And the searchers are so called turtles beginning at random position. When any of the turtle enters the red area, the simulation stops and number of ticks is recorded. Both methods have been done 30 times for 1, 2 and 3 searchers.

==First method - random walk==
In the fist method I implemented random walk. Each step (represented by one tick) one step ahead is made and then the heading is changed by random angle using this formula:
set heading (heading + (angle / 2) - (random angle))

If the searchers hits the border of the forest, he will turn "backwards" as described in this code:
to check-borders
if (xcor < min-pxcor) [set heading (heading + 180)]
if (xcor > max-pxcor) [set heading (heading + 180)]
if (ycor < min-pycor) [set heading (heading + 180)]
if (ycor > max-pycor) [set heading (heading + 180)]
end

[[File:randomwalk_UI.png]]

After measuring 30 times for each angle and 1, 2 and 3 searchers I created table of average steps (ticks) needed to find the searched person. Not agregated data can be found in [[File:Results random walk.xlsx]]

{| class="wikitable"
|+ align="top" | '''Averages for random walk depending on the angle'''
!| Searchers || 45 angle || 90 angle || 180 angle || 360 angle
|-
| 1 || 1033 || 583 || 633 || 1301
|-
| 2 || 398 || 304 || 403 || 658
|-
| 3 || 290 || 219 || 130 || 371
|}

To see how steps needed vary I used standart deviation.

{| class="wikitable"
|+ align="top" | '''Standart deviations for random walk depending on the angle'''
!| Searchers || 45 angle || 90 angle || 180 angle || 360 angle
|-
| 1 || 885 || 543 || 662 || 1793
|-
| 2 || 377 || 236 || 356 || 869
|-
| 3 || 366 || 237 || 128 || 472
|}

==Second method - bouncing==
In the second method searchers and searchers are spawned on the random position. Setup is same as in the first method. However the moving method differs. Searchers are moving in direct way and when they hit border they bounce in oposite direction as described in this code:
to check-bounce
;; bounce off left and right walls
if abs pxcor = max-pxcor [
set heading (- heading)
]
;; bounce off top and bottom walls
if abs pycor = max-pycor [
set heading (180 - heading)
]
end

[[File:bounce_UI.png]]

The method of recording steps needed to find the lost person was same as in the first method, however the angle is not variable now. To see how the data vary I have used standart deviation again. Non agregated data can be seen here [[File: Results bounce.xlsx]]

{| class="wikitable"
|+ align="top" | '''Averages and standart deviations for bouncing method'''
!| Searchers || Average || St. dev.
|-
| 1 || 566 || 624
|-
| 2 || 192 || 161
|-
| 3 || 180 || 141
|}

=Results=
To evaluate random walk method. We can see that if we are searching alone, the 90 angle seems the most efective, however 180 angle seems not bad at all too. If we are searching as two people - 90 angle is the most suitable. When there are three searchers the 180 angle seems the best.

When we want to compare our two methods, For one searcher bounce method is better even then the most effective angle in random walk method. For two searchers it is the same, bounce method is more effective again, even then the 90 angle method. For three searchers is the random walk with the 180 angle most effective.

However if we want to evaluate the standart deviation, basicly it is always better with the bounce method. So it seems like safer choice if we don't want to risk really long journey to find the dog.

=Conclussion=
Although we can see in the tables, that there are differences in both strategies we mustn't forget, that we are using only 30 observation for all the variants. The number should be more significant to say that certainly. However we can see significant difference in standart deviation and thus we can see that using random element in our searching strategy can pay off but also can be pretty bad.

=Code=
==Random walk method==
globals [
found
]

to setup
clear-all
reset-ticks
set found false

setup-searchers
setup-lost-area
end

to setup-searchers
create-turtles searchers [
pen-up
]
ask turtles [
setxy random-xcor random-ycor
]
end

to setup-lost-area
ask patch random-xcor random-ycor [
set pcolor red
]

end

to check-borders
if (xcor < min-pxcor) [set heading (heading + 180)]
if (xcor > max-pxcor) [set heading (heading + 180)]
if (ycor < min-pycor) [set heading (heading + 180)]
if (ycor > max-pycor) [set heading (heading + 180)]
end

to go-random
if found = true [
stop
]
tick
ask turtles [
if pcolor = red [
set found true
]
check-borders
set heading (heading + (angle / 2) - (random angle))
forward 1
]
end

[[File:Sp random walk.nlogo]]

==Bounce method==
globals [
found
]

to setup
clear-all
reset-ticks
set found false

setup-searchers
setup-lost-area
end

to setup-searchers
create-turtles searchers [
pen-up
]
ask turtles [
setxy random-xcor random-ycor
]
end

to setup-lost-area
ask patch random-xcor random-ycor [
set pcolor red
]

end

to check-bounce
;; bounce off left and right walls
if abs pxcor = max-pxcor [
set heading (- heading)
]
;; bounce off top and bottom walls
if abs pycor = max-pycor [
set heading (180 - heading)
]
end

to go
if found = true [
stop
]
tick
ask turtles [
if pcolor = red [
set found true
]
check-bounce
forward 1
]
end

[[File:Sp bounce.nlogo]]

Finding strategies comparison

2023-01-23T18:46:35Z

Kadt02:

This page presents the multi-agent simulation of comparison of strategies for finding a lost person in the forest made by Tomas Kadane.

=Problem definition=
This problem came to my mind because I have once lost my dog in a forest. I was thinking what would be the best strategy to find the dog. However it is pretty complex problem to simulate forest and all types of strategies, so this work is simplified by multiple factors (against the reality and work assignment). I'm comparing two strategies with variable number of searchers in square pane with fixed boundaries.

=Method=
For the simulation I have used NetLogo 6.3.0. I'm implementing two basic searching strategies. First one is based on random walk and the second is based on the walking to the edge of the forest and then bouncing in the opposite direction.

Searched person is represented by single red colored patch (representing area where the searched person can be seen). And the searchers are so called turtles beginning at random position. When any of the turtle enters the red area, the simulation stops and number of ticks is recorded. Both methods have been done 30 times for 1, 2 and 3 searchers.

==First method - random walk==
In the fist method I implemented random walk. Each step (represented by one tick) one step ahead is made and then the heading is changed by random angle using this formula:
set heading (heading + (angle / 2) - (random angle))

If the searchers hits the border of the forest, he will turn "backwards" as described in this code:
to check-borders
if (xcor < min-pxcor) [set heading (heading + 180)]
if (xcor > max-pxcor) [set heading (heading + 180)]
if (ycor < min-pycor) [set heading (heading + 180)]
if (ycor > max-pycor) [set heading (heading + 180)]
end

[[File:randomwalk_UI.png]]

After measuring 30 times for each angle and 1, 2 and 3 searchers I created table of average steps (ticks) needed to find the searched person. Not agregated data can be found in [[File:Results random walk.xlsx]]

{| class="wikitable"
|+ align="top" | '''Averages for random walk depending on the angle'''
!| Searchers || 45 angle || 90 angle || 180 angle || 360 angle
|-
| 1 || 1033 || 583 || 633 || 1301
|-
| 2 || 398 || 304 || 403 || 658
|-
| 3 || 290 || 219 || 130 || 371
|}

To see how steps needed vary I used standart deviation.

{| class="wikitable"
|+ align="top" | '''Standart deviations for random walk depending on the angle'''
!| Searchers || 45 angle || 90 angle || 180 angle || 360 angle
|-
| 1 || 885 || 543 || 662 || 1793
|-
| 2 || 377 || 236 || 356 || 869
|-
| 3 || 366 || 237 || 128 || 472
|}

==Second method - bouncing==
In the second method searchers and searchers are spawned on the random position. Setup is same as in the first method. However the moving method differs. Searchers are moving in direct way and when they hit border they bounce in oposite direction as described in this code:
to check-bounce
;; bounce off left and right walls
if abs pxcor = max-pxcor [
set heading (- heading)
]
;; bounce off top and bottom walls
if abs pycor = max-pycor [
set heading (180 - heading)
]
end

[[File:bounce_UI.png]]

The method of recording steps needed to find the lost person was same as in the first method, however the angle is not variable now. To see how the data vary I have used standart deviation again.

{| class="wikitable"
|+ align="top" | '''Averages and standart deviations for bouncing method'''
!| Searchers || Average || St. dev.
|-
| 1 || 566 || 624
|-
| 2 || 192 || 161
|-
| 3 || 180 || 141
|}

=Results=
To evaluate random walk method. We can see that if we are searching alone, the 90 angle seems the most efective, however 180 angle seems not bad at all too. If we are searching as two people - 90 angle is the most suitable. When there are three searchers the 180 angle seems the best.

When we want to compare our two methods, For one searcher bounce method is better even then the most effective angle in random walk method. For two searchers it is the same, bounce method is more effective again, even then the 90 angle method. For three searchers is the random walk with the 180 angle most effective.

However if we want to evaluate the standart deviation, basicly it is always better with the bounce method. So it seems like safer choice if we don't want to risk really long journey to find the dog.

=Conclussion=
Although we can see in the tables, that there are differences in both strategies we mustn't forget, that we are using only 30 observation for all the variants. The number should be more significant to say that certainly. However we can see significant difference in standart deviation and thus we can see that using random element in our searching strategy can pay off but also can be pretty bad.

=Code=
==Random walk method==
globals [
found
]

to setup
clear-all
reset-ticks
set found false

setup-searchers
setup-lost-area
end

to setup-searchers
create-turtles searchers [
pen-up
]
ask turtles [
setxy random-xcor random-ycor
]
end

to setup-lost-area
ask patch random-xcor random-ycor [
set pcolor red
]

end

to check-borders
if (xcor < min-pxcor) [set heading (heading + 180)]
if (xcor > max-pxcor) [set heading (heading + 180)]
if (ycor < min-pycor) [set heading (heading + 180)]
if (ycor > max-pycor) [set heading (heading + 180)]
end

to go-random
if found = true [
stop
]
tick
ask turtles [
if pcolor = red [
set found true
]
check-borders
set heading (heading + (angle / 2) - (random angle))
forward 1
]
end

[[File:Sp random walk.nlogo]]

==Bounce method==
globals [
found
]

to setup
clear-all
reset-ticks
set found false

setup-searchers
setup-lost-area
end

to setup-searchers
create-turtles searchers [
pen-up
]
ask turtles [
setxy random-xcor random-ycor
]
end

to setup-lost-area
ask patch random-xcor random-ycor [
set pcolor red
]

end

to check-bounce
;; bounce off left and right walls
if abs pxcor = max-pxcor [
set heading (- heading)
]
;; bounce off top and bottom walls
if abs pycor = max-pycor [
set heading (180 - heading)
]
end

to go
if found = true [
stop
]
tick
ask turtles [
if pcolor = red [
set found true
]
check-bounce
forward 1
]
end

[[File:Sp bounce.nlogo]]

WS 2022/2023

2023-01-23T18:44:26Z

Kadt02:

Semestral papers from winter term 2022/2023. Please, put here links to the pages with your paper. First you need to have your [[Assignments WS 2022/2023|assignment approved]]

==Simulations==
--[[User:Julian Bleyer|Julian Bleyer]] ([[User talk:Xkrep33|talk]]) 0:44, 18 January 2023(CET) Aircraft Evacuation Simulation: [[Airplane_Evacuation]]

--[[User:Baumareb|Baumareb]] ([[User talk:Baumareb|talk]]) 13:55, 18 January 2023 (CET) Cartel simulation with leniency program by Rebecca Baumann (baur00): [[cartel_simulation]]

--[[User:Kock06|Kock06]] ([[User talk:Kock06|talk]]) 14:03, 22 January(GTM+8) Receivables prediction by Monte Carlo simulation: [[Receivables]]

--[[User:Ruzv01|Ruzv01]] ([[User talk:Ruzv01|talk]]) 10:33, 22 January(CET) Household electricity consumption: [[Household electricity consumption]]

--[[User:Haki00|Haki00]] ([[User talk:Haki00|talk]]) 21:38, 22 January(CET) Artsakh blockade: [[Artsakh blockade]]

--[[User:Edema|Edema]] ([[User talk:Haki00|talk]]) 22:00, 22 January(CET) Mortgage Assessment: [[Mortgage Assessment]]

--[[User:BortnikSvitlana|BortnikSvitlana]] ([[User talk:BortnikSvitlana|talk]]) 22:36, 22 January(CET) Traffic Simulation at an Intersection: [[Traffic Simulation at an Intersection]]

--[[User:Ceta|Ceta]] ([[User talk:Ceta|talk]]) 22:48, 22 January 2023 (CET) Pumped hydroelectric energy storage (PHES)System Simulation [[Pump_storage]]

--[[User:Botd00|Botd00]] ([[User talk:Botd00|talk]]) 22:46, 22 January(CET) Twitter Simulation: [[Twitter simulation]]

-- [[User:Miln02|Miln02]] ([[User talk:Miln02|talk]]) 23:05, 22 January 2023 (CET) Saving for an apartment [[Savingforanapartment]]

-- [[User:Abizah1|Abizah1]] ([[User talk:Abizah1|talk]]) 23:55, 22 January 2023 (CET) https://www.simulace.info/index.php/Boxing_athlete

-- [[User:Pierreatekwana1|Pierreatekwana1]] ([[User talk:Pierreatekwana1|talk]]) 00:24, 23 January 2023 (CET) Crop yield simulation: [[Cropyield]]

-- [[User:Kane02|Kane02]] ([[User talk:Kane02|talk]]) 18:52, 23 January 2023 (CET) Car Park Solution for a New Cinema [[Car Park Solution for a New Cinema]]

-- [[User:Kadt02|Kadt02]] ([[User talk:Kadt02|talk]]) 19:43, 23 January 2023 (CET) Comparing searching stategies [[Finding strategies comparison]]

Finding strategies comparison

2023-01-23T18:41:34Z

Kadt02: Created page with "This page presents the work of Comparison of strategies for finding a lost person in the forest made by Tomas Kadane =Problem definition= This problem came to my mind because..."

This page presents the work of Comparison of strategies for finding a lost person in the forest made by Tomas Kadane

=Problem definition=
This problem came to my mind because I have once lost my dog in a forest. I was thinking what would be the best strategy to find the dog. However it is pretty complex problem to simulate forest and all types of strategies, so this work is simplified by multiple factors (against the reality and work assignment). I'm comparing two strategies with variable number of searchers in square pane with fixed boundaries.

=Method=
For the simulation I have used NetLogo 6.3.0. I'm implementing two basic searching strategies. First one is based on random walk and the second is based on the walking to the edge of the forest and then bouncing in the opposite direction.

Searched person is represented by single red colored patch (representing area where the searched person can be seen). And the searchers are so called turtles beginning at random position. When any of the turtle enters the red area, the simulation stops and number of ticks is recorded. Both methods have been done 30 times for 1, 2 and 3 searchers.

==First method - random walk==
In the fist method I implemented random walk. Each step (represented by one tick) one step ahead is made and then the heading is changed by random angle using this formula:
set heading (heading + (angle / 2) - (random angle))

If the searchers hits the border of the forest, he will turn "backwards" as described in this code:
to check-borders
if (xcor < min-pxcor) [set heading (heading + 180)]
if (xcor > max-pxcor) [set heading (heading + 180)]
if (ycor < min-pycor) [set heading (heading + 180)]
if (ycor > max-pycor) [set heading (heading + 180)]
end

[[File:randomwalk_UI.png]]

After measuring 30 times for each angle and 1, 2 and 3 searchers I created table of average steps (ticks) needed to find the searched person. Not agregated data can be found in [[File:Results random walk.xlsx]]

{| class="wikitable"
|+ align="top" | '''Averages for random walk depending on the angle'''
!| Searchers || 45 angle || 90 angle || 180 angle || 360 angle
|-
| 1 || 1033 || 583 || 633 || 1301
|-
| 2 || 398 || 304 || 403 || 658
|-
| 3 || 290 || 219 || 130 || 371
|}

To see how steps needed vary I used standart deviation.

{| class="wikitable"
|+ align="top" | '''Standart deviations for random walk depending on the angle'''
!| Searchers || 45 angle || 90 angle || 180 angle || 360 angle
|-
| 1 || 885 || 543 || 662 || 1793
|-
| 2 || 377 || 236 || 356 || 869
|-
| 3 || 366 || 237 || 128 || 472
|}

==Second method - bouncing==
In the second method searchers and searchers are spawned on the random position. Setup is same as in the first method. However the moving method differs. Searchers are moving in direct way and when they hit border they bounce in oposite direction as described in this code:
to check-bounce
;; bounce off left and right walls
if abs pxcor = max-pxcor [
set heading (- heading)
]
;; bounce off top and bottom walls
if abs pycor = max-pycor [
set heading (180 - heading)
]
end

[[File:bounce_UI.png]]

The method of recording steps needed to find the lost person was same as in the first method, however the angle is not variable now. To see how the data vary I have used standart deviation again.

{| class="wikitable"
|+ align="top" | '''Averages and standart deviations for bouncing method'''
!| Searchers || Average || St. dev.
|-
| 1 || 566 || 624
|-
| 2 || 192 || 161
|-
| 3 || 180 || 141
|}

=Results=
To evaluate random walk method. We can see that if we are searching alone, the 90 angle seems the most efective, however 180 angle seems not bad at all too. If we are searching as two people - 90 angle is the most suitable. When there are three searchers the 180 angle seems the best.

When we want to compare our two methods, For one searcher bounce method is better even then the most effective angle in random walk method. For two searchers it is the same, bounce method is more effective again, even then the 90 angle method. For three searchers is the random walk with the 180 angle most effective.

However if we want to evaluate the standart deviation, basicly it is always better with the bounce method. So it seems like safer choice if we don't want to risk really long journey to find the dog.

=Conclussion=
Although we can see in the tables, that there are differences in both strategies we mustn't forget, that we are using only 30 observation for all the variants. The number should be more significant to say that certainly. However we can see significant difference in standart deviation and thus we can see that using random element in our searching strategy can pay off but also can be pretty bad.

=Code=
==Random walk method==
globals [
found
]

to setup
clear-all
reset-ticks
set found false

setup-searchers
setup-lost-area
end

to setup-searchers
create-turtles searchers [
pen-up
]
ask turtles [
setxy random-xcor random-ycor
]
end

to setup-lost-area
ask patch random-xcor random-ycor [
set pcolor red
]

end

to check-borders
if (xcor < min-pxcor) [set heading (heading + 180)]
if (xcor > max-pxcor) [set heading (heading + 180)]
if (ycor < min-pycor) [set heading (heading + 180)]
if (ycor > max-pycor) [set heading (heading + 180)]
end

to go-random
if found = true [
stop
]
tick
ask turtles [
if pcolor = red [
set found true
]
check-borders
set heading (heading + (angle / 2) - (random angle))
forward 1
]
end

[[File:Sp random walk.nlogo]]

==Bounce method==
globals [
found
]

to setup
clear-all
reset-ticks
set found false

setup-searchers
setup-lost-area
end

to setup-searchers
create-turtles searchers [
pen-up
]
ask turtles [
setxy random-xcor random-ycor
]
end

to setup-lost-area
ask patch random-xcor random-ycor [
set pcolor red
]

end

to check-bounce
;; bounce off left and right walls
if abs pxcor = max-pxcor [
set heading (- heading)
]
;; bounce off top and bottom walls
if abs pycor = max-pycor [
set heading (180 - heading)
]
end

to go
if found = true [
stop
]
tick
ask turtles [
if pcolor = red [
set found true
]
check-bounce
forward 1
]
end

[[File:Sp bounce.nlogo]]

File:Bounce UI.png

2023-01-23T18:25:42Z

Kadt02:

File:Sp random walk.nlogo

2023-01-23T18:10:03Z

Kadt02:

File:Sp bounce.nlogo

2023-01-23T18:09:57Z

Kadt02:

File:Results random walk.xlsx

2023-01-23T18:09:24Z

Kadt02:

File:Results bounce.xlsx

2023-01-23T18:08:52Z

Kadt02:

File:Randomwalk UI.png

2023-01-23T18:02:07Z

Kadt02:

Assignments WS 2022/2023

2022-12-17T15:01:35Z

Kadt02:

{{Ambox
| text = <div>
Please, put here your assignments. Do not forget to sign them. You can use <nowiki>~~~~</nowiki> (four tildas) for an automatic signature. Use Show preview in order to check the result before your final sumbition.
</div>
}}

{{Ambox
| text = <div>
Please, strive to formulate your assignment carefully. We expect an adequate effort to formulate the assignment as it is your semestral paper. Do not forget that your main goal is a research paper. It means your simulation model must generate the results that are specific, measurable and verifiable. Think twice how you will develop your model, which entities you will use, draw a model diagram, consider what you will measure. No sooner than when you have a good idea about the model, submit your assignment. And of course, read [[How to deal with the simulation assignment|How to deal with the simulation assignment]].
</div>
}}

{{Ambox
| text = <div>
Topics on gambling, cards, etc. are not welcome.
</div>
}}

{{Ambox
| type = content
| text = <div>
In order to avoid possible confusion, please, check if you have added '''approved''' in bold somewhere in our comment under your submission. If there is no '''approved''', it means the assignment was not approved yet.
</div>
}}

== Effect of leniency programs on cartel rates by [[User:Baumareb|Baumareb]] ([[User talk:Baumareb|talk]]) 11:18, 7 December 2022 (CET) ==

''' Simulation '''

The leniency program of the European Commission offers the companies involved in a cartel either complete or partial immunity from fines if they self-report and hand over evidence. It was introduced in 1996, following the surge in amnesty applications in the wake of the 1993 revision of the Corporate Leniency Program of the US Department of Justice’s Antitrust Division. Reports from various implemented leniency programs showed that such programs led to numerous applications. However, despite the clear increase in leniency applications, the question poses itself as to whether the programs were also successful in a sense that the actual cartel rate in those countries declined.
The simulation will be based on a study of Harrington and Chang from 2015, in which they concluded the following:

• The actual cartel rate decreases in case that the leniency program does not affect the non-leniency enforcement

• But: if the non-leniency enforcement is affected because resources are shifted to the prosecution of leniency application cases, there might be two possibilities, the cartel rate might increase.

This simulation focuses on the latter case. Assuming endogenized non-leniency enforcement, the introduction of a leniency program might have a differential impact on different industries. If a leniency program is introduced, the cartels that are about to collapse will seek to self-report. This in turn shifts resources from exposing active cartels to prosecuting cartels that are already collapsing. This creates more work for the authorities, who, instead of focusing on active cartels may now focus on dying cartels. This crowding-out effect coming about with the introduction of a leniency program shall be simulated in this project.

''' Goal '''

The simulation will have the following objectives:

* Illustrate the change in cartel rates and the change in the average life expectancy of a cartel triggered by the introduction of a leniency program in case of endogenized non-leniency enforcement for industries with unstable cartels (e.g. industries with a high number of competitors, or demand with more price elasticity) and for industries with stable cartels (e.g. industries with less competitors and demand with less price elasticity).
* Illustrate how many resources may be shifted from non-leniency enforcement to prosecuting leniency application cases without it having an undesired effect on the actual cartel rate.

''' Practical relevance '''

The simulation may be used by law enforcement officials to evaluate whether a leniency program leads to the desired effect (i.e. the decrease in the cartel rate) or not. Also, it can help for deciding whether the non-leniency enforcement needs to be strengthened to prevent the crowding-out effect.

''' Method '''

The described scenario is a multi-agent simulation in which the agents are pursuing a utility-based approach. Thus, the simulation will be done with NetLogo.
The following features will be included into the simulation:

- For both industries with stable and industries with unstable cartels:

* Number of active cartels (dying after reaching avg. life expectancy)
* Number of competitors
* Average life expectancy of a cartel
* “Birth” of new cartels

- For leniency/non-leniency enforcement:
* Resources and their assignment to either leniency or non-leniency enforcement
* Capacity of taking down an active cartel
* Capacity of taking down a cartel based on leniency applications

The simulation will be based on the 2015 research from Harrington and Chang as well as on publicly accessible data from the European Commission regarding antitrust cases from 1964 until today.

''' Sources '''
* Harrington Jr, J. E., & Chang, M. H. (2015). When can we expect a corporate leniency program to result in fewer cartels?. The Journal of Law and Economics, 58(2), 417-449.
* Ordóñez‐De‐Haro, J. M., Borrell, J. R., & Jiménez, J. L. (2018). The European commission's fight against cartels (1962–2014): A retrospective and forensic analysis. JCMS: Journal of Common Market Studies, 56(5), 1087-1107.

[[User:Baumareb|Baumareb]] ([[User talk:Baumareb|talk]]) 11:18, 7 December 2022 (CET) Rebecca Baumann (baur00)

: This isn't an easy topic. Be careful about available data. '''Approved''' [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 01:46, 15 December 2022 (CET)

== The prediction of divorce rate in Czech Republic for the following 50 years ==

''' The goal of the simulation '''

Divorce in the Czech republic must always contain at least one hearing in front of the court. Legally, there are many more parties involved, such as a notary, who must verify the signatures on all the important documents and many times, divorce lawyers are also necessary. To be able to satisfy the needs of the public, all the involved parties must have an idea about how many married couples are likely to get divorced in the years to come. This simulation will help prepare the courts, notaries and lawyers by making a prediction on the amount of divorces in the next 50 years. This will also help law students choose the field of law that they will specialize in by answering the question whether divorce lawyers will be necessary in the future or not.

''' Method '''

Vensim will be used for this simulation. The used data will come from the Czech Statistical Office and possibly other sources (Refer to [1] and [2]), such as published studies on the most common reasons for divorce. When possible, the data about each reason of divorce will be also found and the simulation model will contain this data.

== Edit: additional details ==

''' What all parameters will the simulation work with and how?'''

1. Number of marriages – the more marriages, the more divorces

a/ Number of people in the age 25 to 34 (i.e., the most common age to get married) – the more there is of these people, the more marriages there will be

b/ Number of divorced people in the age 40 to 49 (i.e., the most common age to get re-married after a divorce) – the more there is of these people, the more marriages there will be, however not as much as the number above

2. Micro causes of divorces = Top 10 causes of divorce as researched by the Czech Statistical Office, published yearly – the more common are these causes (alcoholism, infidelity etc), the more divorces there will be

a/ Ill-considered marriage

b/ Alcoholism

c/ Infidelity

d/ Lack of interest in the family (incl. abandon. of living together)

e/ Ill-treatment, criminal conviction

f/ Different characters, views and interests

g/ Health reasons

h/ Sexual discord

i/ Other causes

j/ Cause not given

3. Number of people in the age 40 to 49 – the more there is of these people, the more divorces there will be (it is the most common age to get divorced)

4. Macro causes of divorces

a/ Economic independence of women = the more economically independent women are, the more likely they are to divorce in case of an unhappy marriage – this will be evaluated through a comparison of data of average income of men vs. women

b/ Being religious – divorce is far less common for religious people.

''' What data source will be used for deriving the equations?'''

Based on my current research of data sources, the Czech Statistical Office has the all the data necessary for this paper.

[1] Scott, S. B., Rhoades, G. K., Stanley, S. M., Allen, E. S., & Markman, H. J. (2013). Reasons for Divorce and Recollections of Premarital Intervention: Implications for Improving Relationship Education. Couple & family psychology, 2(2), 131–145. https://doi.org/10.1037/a0032025

[2] Hawkins, Alan & Willoughby, Brian & Doherty, William. (2012). Reasons for Divorce and Openness to Marital Reconciliation. Journal of Divorce & Remarriage. 53. 453-463. 10.1080/10502556.2012.682898.

: Sounds interesting, but I miss more detail about the simulation. What all parameters will the simulation work with and how? What data source will be used for deriving the equations? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 11:03, 15 December 2022 (CET)
:: ''' Approved'''. Just make sure that the equtions, reasons for divorce and their impact on divorce rate are properly quantified.[[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 07:23, 17 December 2022 (CET)

==Crop Yield Forecasting==

''' Simulation '''

Crop growth and development simulations and yield forecasting will be performed using variables such as crop type, planting date, soil type, soil texture, and climate data (temperature, rainfall, etc.).

'''Problem definition'''

Arable land is increasingly limited, while the world's population has steadily been increasing over the years. In order to meet rapidly rising demand, production must be increased while natural resources must be protected. New agricultural research is needed to provide information on how to achieve sustainable agriculture in the face of global climate variability. Predicting crop yield under different conditions, such as different irrigation regimes, planting dates, and crop management practices, has become critical for farmers and other stakeholders who use these predictions to make more informed decisions about how to allocate resources, such as labor, equipment, and inputs, to maximize yield and productivity.

'''Method'''

Crop yield simulation tools include AquaCrop, DSSAT, and CropSyst. These tools use mathematical models to simulate crop growth and development based on input data like weather, soil type, and management practices. These tools use this data to estimate the crop's potential yield, as well as other important factors like water use and crop evapotranspiration. For this assignment I will be using AquaCrop which is a crop water productivity model developed by the United Nations Food and Agriculture Organization (FAO). It is used to simulate crop growth and yield under various environmental and management conditions. AquaCrop simulates crop growth and development, and estimates yield based on soil conditions, climate, irrigation, and management practices. The application gives access to various FAO databases with all the necessary data needed to perform a comprehensive simulation of the crop yield.

'''Citations'''

* Y. Lu, C. Wei, M. F. McCabe, and J. Sheffield, “Multi-variable assimilation into a modified AquaCrop model for improved maize simulation without management or crop phenology information,” Agricultural Water Management, vol. 266, p. 107576, May 2022, doi: 10.1016/j.agwat.2022.107576.
* P. N. Kephe, K. K. Ayisi, and B. M. Petja, “Challenges and opportunities in crop simulation modelling under seasonal and projected climate change scenarios for crop production in South Africa,” Agriculture & Food Security, vol. 10, no. 1, p. 10, Apr. 2021, doi: 10.1186/s40066-020-00283-5.
* N. T. Olivera, O. B. Manrique, Y. G. Masjuan, and A. M. G. Alega, “Evaluation of AquaCrop model in crop dry bean growth simulation,” Revista Ciencias Técnicas Agropecuarias, vol. 25, no. 3, pp. 23–30, Accessed: Dec. 10, 2022. [Online]. Available: https://www.redalyc.org/journal/932/93246970003/html/
* N. Pirmoradian, Z. Saadati, M. Rezaei, and M. R. Khaledian, “Simulating water productivity of paddy rice under irrigation regimes using AquaCrop model in humid and semiarid regions of Iran,” Appl Water Sci, vol. 10, no. 7, p. 161, Jun. 2020, doi: 10.1007/s13201-020-01249-5.

[[User:Pierreatekwana|Pierreatekwana]] ([[User talk:Pierreatekwana|talk]]) 15:06, 15 December 2022 (CET)

: Topic souds interesting, but the proposed simulation tool has to be one of the ones we have used in our class ( as specified in How to deal with the simulation assignment:
One of your key course requirements is a submission of simulation. You choose your topic yourself, the same as a method and a tool that you will use. It could be any of the development environments we have used (Excel, Simprocess, Netlogo, or Vensim).) [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 07:05, 17 December 2022 (CET)

== Electricity Spot Market Simulation by [[User:Ceta|Ceta]] ([[User talk:Ceta|talk]]) 01:13, 16 December 2022 (CET) ==

'''Simulation'''

I’ve been working as a professional in the Turkish Electricity Market (EXIST) for more than 10 years. I had the chance to practice in business some fundamental methods such as Stochastic Dual Dynamic Programming (SDDP) for Hydro-Thermal dispatch optimization. The spot exchange of power or the power market itself has many examples of simulation. And the simulation of the uncertainties both in the short term and the long term are very critical in terms of new investments, portfolio management, and resource optimization.

'''Method'''

Both Excel and Netlogo simulations can be derived

'''• Inputs
'''

- Portfolios

(Thermals: Coal-Fired, Natural Gas-Fired, Other Thermals; Renewables: Hydro, Wind, Solar PV, Other Renewables)

- Resource Prices (Eur/mwh)

- Installed Capacity (in MW)

'''• Variables'''

- Power Demand (MWh)

- Calendar Day

'''• Uncertain Events
'''
- Maintenance (availability)

- Climate Factors (Wind, Hydro Sources, Temperature effect on Load-Demand)

- Natural Gas Shortage (Due to Climate Factors and/or Supply)

'''• Outputs
'''
- Marginal Cost

- Market Clearing Prices (MCP)

•''' Goals
'''
- Objective Function Y= Min (MCP)

- Effects of Natural Gas Shortages

- Effects of Climate Factors

- Effects of Maintenance

- Effects of Calendar

- Sensitivity Analysis of commissioning a Nuclear Power Plant in the market

'''Citations'''

- Sensfuß, Frank; Ragwitz, Mario; Genoese, Massimo; Möst, Dominik (2007) : Agent-based simulation of electricity markets: a literature review, Working Paper Sustainability and Innovation, No. S5/2007, Fraunhofer-Institut für System- und Innovationsforschung ISI, Karlsruhe,

- https://nbn-resolving.de/urn:nbn:de:0011-n-661574

- https://www.econstor.eu/bitstream/10419/28520/1/570113083.pdf

- https://www.tudelft.nl/evenementen/2021/powerweb/electricity-market-simulation-game ==

''' Data '''

- EXIST Transparency Portal https://seffaflik.epias.com.tr/transparency/

- Electricity Transmission Operator https://www.teias.gov.tr/en-US

== Profit in store vs e-shop ==

''' Method:''' System Dynamics

'''Software:''' Vensim

'''Simulation'''

An unnamed company that sells carpets has its own store in Prague. During COVID-19 the company reopened an e-shop, so it currently has two mutually supporting sales channels. Both types of stores have their advantages and disadvantages. At the same time, there are various factors that affect the profit. Examples of these factors are the following: customer satisfaction and needs (carpet quality, order processing speed, price, etc.), expenses (advertising, rent, employees, etc.), the possibility of expansion, etc.

'''Model parameters'''

*Expenses
**fixed
**variable
*Revenues
**customer satisfaction
**price
**a number of sales, etc.

''' The goal of the simulation '''

The goal of this simulation is to find out how the company's profits will develop in the next 5 years, individually in the store and e-shop. Find out what parameters can increase profits the most, individually for each type of store, and compare these parameters.

'''Data'''

Real data provided by the owners of the store

[[User:Ploo00|Ploo00]] ([[User talk:Ploo00|talk]]) 01:41, 16 December 2022 (CET)
:Please elaborate in more detail as we have discussed in class [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 07:17, 17 December 2022 (CET)

<nowiki>~~~~</nowiki>

==Comparison of strategies for finding a lost person in the forest==

''' Author:''' Tomáš Kadaně (kadt02)

'''Type:''' Multi-agent

'''Software:''' NetLogo

'''Description:'''

The simulation will focus on comparing the times needed to find a lost person in a forest (area with trees).
The metric to compare the strategies will be the number of ticks needed to find the wanted person. Both the person being searched for and the searcher will be in a random location at the beginning of the simulation.
Within the simulation, I will take several measurements for each strategy and number of searchers (1 to 5), so that the number is statistically significant and use, for example, the means to compare which strategy is the most appropriate.

The model will be able to simulate several search strategies

*one step forward and then turn of random degree (-45 to 45 degrees), so random walk
*walk straight until it hits the edge of the forest or tree, then turn and continue walking straight
*first walk to the nearest corner of the forest and then a some kind of serpentine search
*possibly other strategies

'''Goals:'''

Finding the most appropriate strategy for finding a person in the forest depending on the number of people searching.

'''Agents:'''

*Searchers (e.g. police officers)
*Lost person

'''Parameters:'''

*Number of searchers
*Type of strategy
*Ticks needed to find person

'''Possible extensions:'''

*Searchers with certain pace of walking
*Finding the person won’t mean be at same location but seeing it for some distance (again certain ability of the searcher to see for certain distance)
*Cooperation of finders (formations, place distribution)
*Lost person will be moving when being looked for

[[User:Kadt02|Kadt02]] ([[User talk:Kadt02|talk]]) 16:01, 17 December 2022 (CET)

Assignments WS 2022/2023

2022-12-17T15:00:28Z

Kadt02:

Assignments WS 2022/2023

2022-12-17T14:59:45Z

Kadt02:

{{Ambox
| text = <div>
Please, put here your assignments. Do not forget to sign them. You can use <nowiki>~~~~</nowiki> (four tildas) for an automatic signature. Use Show preview in order to check the result before your final sumbition.
</div>
}}

{{Ambox
| text = <div>
Please, strive to formulate your assignment carefully. We expect an adequate effort to formulate the assignment as it is your semestral paper. Do not forget that your main goal is a research paper. It means your simulation model must generate the results that are specific, measurable and verifiable. Think twice how you will develop your model, which entities you will use, draw a model diagram, consider what you will measure. No sooner than when you have a good idea about the model, submit your assignment. And of course, read [[How to deal with the simulation assignment|How to deal with the simulation assignment]].
</div>
}}

{{Ambox
| text = <div>
Topics on gambling, cards, etc. are not welcome.
</div>
}}

{{Ambox
| type = content
| text = <div>
In order to avoid possible confusion, please, check if you have added '''approved''' in bold somewhere in our comment under your submission. If there is no '''approved''', it means the assignment was not approved yet.
</div>
}}

== Effect of leniency programs on cartel rates by [[User:Baumareb|Baumareb]] ([[User talk:Baumareb|talk]]) 11:18, 7 December 2022 (CET) ==

''' Simulation '''

The leniency program of the European Commission offers the companies involved in a cartel either complete or partial immunity from fines if they self-report and hand over evidence. It was introduced in 1996, following the surge in amnesty applications in the wake of the 1993 revision of the Corporate Leniency Program of the US Department of Justice’s Antitrust Division. Reports from various implemented leniency programs showed that such programs led to numerous applications. However, despite the clear increase in leniency applications, the question poses itself as to whether the programs were also successful in a sense that the actual cartel rate in those countries declined.
The simulation will be based on a study of Harrington and Chang from 2015, in which they concluded the following:

• The actual cartel rate decreases in case that the leniency program does not affect the non-leniency enforcement

• But: if the non-leniency enforcement is affected because resources are shifted to the prosecution of leniency application cases, there might be two possibilities, the cartel rate might increase.

This simulation focuses on the latter case. Assuming endogenized non-leniency enforcement, the introduction of a leniency program might have a differential impact on different industries. If a leniency program is introduced, the cartels that are about to collapse will seek to self-report. This in turn shifts resources from exposing active cartels to prosecuting cartels that are already collapsing. This creates more work for the authorities, who, instead of focusing on active cartels may now focus on dying cartels. This crowding-out effect coming about with the introduction of a leniency program shall be simulated in this project.

''' Goal '''

The simulation will have the following objectives:

* Illustrate the change in cartel rates and the change in the average life expectancy of a cartel triggered by the introduction of a leniency program in case of endogenized non-leniency enforcement for industries with unstable cartels (e.g. industries with a high number of competitors, or demand with more price elasticity) and for industries with stable cartels (e.g. industries with less competitors and demand with less price elasticity).
* Illustrate how many resources may be shifted from non-leniency enforcement to prosecuting leniency application cases without it having an undesired effect on the actual cartel rate.

''' Practical relevance '''

The simulation may be used by law enforcement officials to evaluate whether a leniency program leads to the desired effect (i.e. the decrease in the cartel rate) or not. Also, it can help for deciding whether the non-leniency enforcement needs to be strengthened to prevent the crowding-out effect.

''' Method '''

The described scenario is a multi-agent simulation in which the agents are pursuing a utility-based approach. Thus, the simulation will be done with NetLogo.
The following features will be included into the simulation:

- For both industries with stable and industries with unstable cartels:

* Number of active cartels (dying after reaching avg. life expectancy)
* Number of competitors
* Average life expectancy of a cartel
* “Birth” of new cartels

- For leniency/non-leniency enforcement:
* Resources and their assignment to either leniency or non-leniency enforcement
* Capacity of taking down an active cartel
* Capacity of taking down a cartel based on leniency applications

The simulation will be based on the 2015 research from Harrington and Chang as well as on publicly accessible data from the European Commission regarding antitrust cases from 1964 until today.

''' Sources '''
* Harrington Jr, J. E., & Chang, M. H. (2015). When can we expect a corporate leniency program to result in fewer cartels?. The Journal of Law and Economics, 58(2), 417-449.
* Ordóñez‐De‐Haro, J. M., Borrell, J. R., & Jiménez, J. L. (2018). The European commission's fight against cartels (1962–2014): A retrospective and forensic analysis. JCMS: Journal of Common Market Studies, 56(5), 1087-1107.

[[User:Baumareb|Baumareb]] ([[User talk:Baumareb|talk]]) 11:18, 7 December 2022 (CET) Rebecca Baumann (baur00)

: This isn't an easy topic. Be careful about available data. '''Approved''' [[User:Tomáš|Tomáš]] ([[User talk:Tomáš|talk]]) 01:46, 15 December 2022 (CET)

== The prediction of divorce rate in Czech Republic for the following 50 years ==

''' The goal of the simulation '''

Divorce in the Czech republic must always contain at least one hearing in front of the court. Legally, there are many more parties involved, such as a notary, who must verify the signatures on all the important documents and many times, divorce lawyers are also necessary. To be able to satisfy the needs of the public, all the involved parties must have an idea about how many married couples are likely to get divorced in the years to come. This simulation will help prepare the courts, notaries and lawyers by making a prediction on the amount of divorces in the next 50 years. This will also help law students choose the field of law that they will specialize in by answering the question whether divorce lawyers will be necessary in the future or not.

''' Method '''

Vensim will be used for this simulation. The used data will come from the Czech Statistical Office and possibly other sources (Refer to [1] and [2]), such as published studies on the most common reasons for divorce. When possible, the data about each reason of divorce will be also found and the simulation model will contain this data.

== Edit: additional details ==

''' What all parameters will the simulation work with and how?'''

1. Number of marriages – the more marriages, the more divorces

a/ Number of people in the age 25 to 34 (i.e., the most common age to get married) – the more there is of these people, the more marriages there will be

b/ Number of divorced people in the age 40 to 49 (i.e., the most common age to get re-married after a divorce) – the more there is of these people, the more marriages there will be, however not as much as the number above

2. Micro causes of divorces = Top 10 causes of divorce as researched by the Czech Statistical Office, published yearly – the more common are these causes (alcoholism, infidelity etc), the more divorces there will be

a/ Ill-considered marriage

b/ Alcoholism

c/ Infidelity

d/ Lack of interest in the family (incl. abandon. of living together)

e/ Ill-treatment, criminal conviction

f/ Different characters, views and interests

g/ Health reasons

h/ Sexual discord

i/ Other causes

j/ Cause not given

3. Number of people in the age 40 to 49 – the more there is of these people, the more divorces there will be (it is the most common age to get divorced)

4. Macro causes of divorces

a/ Economic independence of women = the more economically independent women are, the more likely they are to divorce in case of an unhappy marriage – this will be evaluated through a comparison of data of average income of men vs. women

b/ Being religious – divorce is far less common for religious people.

''' What data source will be used for deriving the equations?'''

Based on my current research of data sources, the Czech Statistical Office has the all the data necessary for this paper.

[1] Scott, S. B., Rhoades, G. K., Stanley, S. M., Allen, E. S., & Markman, H. J. (2013). Reasons for Divorce and Recollections of Premarital Intervention: Implications for Improving Relationship Education. Couple & family psychology, 2(2), 131–145. https://doi.org/10.1037/a0032025

[2] Hawkins, Alan & Willoughby, Brian & Doherty, William. (2012). Reasons for Divorce and Openness to Marital Reconciliation. Journal of Divorce & Remarriage. 53. 453-463. 10.1080/10502556.2012.682898.

: Sounds interesting, but I miss more detail about the simulation. What all parameters will the simulation work with and how? What data source will be used for deriving the equations? [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 11:03, 15 December 2022 (CET)
:: ''' Approved'''. Just make sure that the equtions, reasons for divorce and their impact on divorce rate are properly quantified.[[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 07:23, 17 December 2022 (CET)

==Crop Yield Forecasting==

''' Simulation '''

Crop growth and development simulations and yield forecasting will be performed using variables such as crop type, planting date, soil type, soil texture, and climate data (temperature, rainfall, etc.).

'''Problem definition'''

Arable land is increasingly limited, while the world's population has steadily been increasing over the years. In order to meet rapidly rising demand, production must be increased while natural resources must be protected. New agricultural research is needed to provide information on how to achieve sustainable agriculture in the face of global climate variability. Predicting crop yield under different conditions, such as different irrigation regimes, planting dates, and crop management practices, has become critical for farmers and other stakeholders who use these predictions to make more informed decisions about how to allocate resources, such as labor, equipment, and inputs, to maximize yield and productivity.

'''Method'''

Crop yield simulation tools include AquaCrop, DSSAT, and CropSyst. These tools use mathematical models to simulate crop growth and development based on input data like weather, soil type, and management practices. These tools use this data to estimate the crop's potential yield, as well as other important factors like water use and crop evapotranspiration. For this assignment I will be using AquaCrop which is a crop water productivity model developed by the United Nations Food and Agriculture Organization (FAO). It is used to simulate crop growth and yield under various environmental and management conditions. AquaCrop simulates crop growth and development, and estimates yield based on soil conditions, climate, irrigation, and management practices. The application gives access to various FAO databases with all the necessary data needed to perform a comprehensive simulation of the crop yield.

'''Citations'''

* Y. Lu, C. Wei, M. F. McCabe, and J. Sheffield, “Multi-variable assimilation into a modified AquaCrop model for improved maize simulation without management or crop phenology information,” Agricultural Water Management, vol. 266, p. 107576, May 2022, doi: 10.1016/j.agwat.2022.107576.
* P. N. Kephe, K. K. Ayisi, and B. M. Petja, “Challenges and opportunities in crop simulation modelling under seasonal and projected climate change scenarios for crop production in South Africa,” Agriculture & Food Security, vol. 10, no. 1, p. 10, Apr. 2021, doi: 10.1186/s40066-020-00283-5.
* N. T. Olivera, O. B. Manrique, Y. G. Masjuan, and A. M. G. Alega, “Evaluation of AquaCrop model in crop dry bean growth simulation,” Revista Ciencias Técnicas Agropecuarias, vol. 25, no. 3, pp. 23–30, Accessed: Dec. 10, 2022. [Online]. Available: https://www.redalyc.org/journal/932/93246970003/html/
* N. Pirmoradian, Z. Saadati, M. Rezaei, and M. R. Khaledian, “Simulating water productivity of paddy rice under irrigation regimes using AquaCrop model in humid and semiarid regions of Iran,” Appl Water Sci, vol. 10, no. 7, p. 161, Jun. 2020, doi: 10.1007/s13201-020-01249-5.

[[User:Pierreatekwana|Pierreatekwana]] ([[User talk:Pierreatekwana|talk]]) 15:06, 15 December 2022 (CET)

: Topic souds interesting, but the proposed simulation tool has to be one of the ones we have used in our class ( as specified in How to deal with the simulation assignment:
One of your key course requirements is a submission of simulation. You choose your topic yourself, the same as a method and a tool that you will use. It could be any of the development environments we have used (Excel, Simprocess, Netlogo, or Vensim).) [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 07:05, 17 December 2022 (CET)

== Electricity Spot Market Simulation by [[User:Ceta|Ceta]] ([[User talk:Ceta|talk]]) 01:13, 16 December 2022 (CET) ==

'''Simulation'''

I’ve been working as a professional in the Turkish Electricity Market (EXIST) for more than 10 years. I had the chance to practice in business some fundamental methods such as Stochastic Dual Dynamic Programming (SDDP) for Hydro-Thermal dispatch optimization. The spot exchange of power or the power market itself has many examples of simulation. And the simulation of the uncertainties both in the short term and the long term are very critical in terms of new investments, portfolio management, and resource optimization.

'''Method'''

Both Excel and Netlogo simulations can be derived

'''• Inputs
'''

- Portfolios

(Thermals: Coal-Fired, Natural Gas-Fired, Other Thermals; Renewables: Hydro, Wind, Solar PV, Other Renewables)

- Resource Prices (Eur/mwh)

- Installed Capacity (in MW)

'''• Variables'''

- Power Demand (MWh)

- Calendar Day

'''• Uncertain Events
'''
- Maintenance (availability)

- Climate Factors (Wind, Hydro Sources, Temperature effect on Load-Demand)

- Natural Gas Shortage (Due to Climate Factors and/or Supply)

'''• Outputs
'''
- Marginal Cost

- Market Clearing Prices (MCP)

•''' Goals
'''
- Objective Function Y= Min (MCP)

- Effects of Natural Gas Shortages

- Effects of Climate Factors

- Effects of Maintenance

- Effects of Calendar

- Sensitivity Analysis of commissioning a Nuclear Power Plant in the market

'''Citations'''

- Sensfuß, Frank; Ragwitz, Mario; Genoese, Massimo; Möst, Dominik (2007) : Agent-based simulation of electricity markets: a literature review, Working Paper Sustainability and Innovation, No. S5/2007, Fraunhofer-Institut für System- und Innovationsforschung ISI, Karlsruhe,

- https://nbn-resolving.de/urn:nbn:de:0011-n-661574

- https://www.econstor.eu/bitstream/10419/28520/1/570113083.pdf

- https://www.tudelft.nl/evenementen/2021/powerweb/electricity-market-simulation-game ==

''' Data '''

- EXIST Transparency Portal https://seffaflik.epias.com.tr/transparency/

- Electricity Transmission Operator https://www.teias.gov.tr/en-US

== Profit in store vs e-shop ==

''' Method:''' System Dynamics

'''Software:''' Vensim

'''Simulation'''

An unnamed company that sells carpets has its own store in Prague. During COVID-19 the company reopened an e-shop, so it currently has two mutually supporting sales channels. Both types of stores have their advantages and disadvantages. At the same time, there are various factors that affect the profit. Examples of these factors are the following: customer satisfaction and needs (carpet quality, order processing speed, price, etc.), expenses (advertising, rent, employees, etc.), the possibility of expansion, etc.

'''Model parameters'''

*Expenses
**fixed
**variable
*Revenues
**customer satisfaction
**price
**a number of sales, etc.

''' The goal of the simulation '''

The goal of this simulation is to find out how the company's profits will develop in the next 5 years, individually in the store and e-shop. Find out what parameters can increase profits the most, individually for each type of store, and compare these parameters.

'''Data'''

Real data provided by the owners of the store

[[User:Ploo00|Ploo00]] ([[User talk:Ploo00|talk]]) 01:41, 16 December 2022 (CET)
:Please elaborate in more detail as we have discussed in class [[User:Oleg.Svatos|Oleg.Svatos]] ([[User talk:Oleg.Svatos|talk]]) 07:17, 17 December 2022 (CET)

<nowiki>~~~~</nowiki>

==Comparison of strategies for finding a lost person in the forest==

''' Author:''' Tomáš Kadaně (kadt02)

'''Type:''' Multi-agent

'''Software:''' Netlogo

'''Description:'''

The simulation will focus on comparing the times needed to find a lost person in a forest (area with trees).
The metric to compare the strategies will be the number of ticks needed to find the wanted person. Both the person being searched for and the searcher will be in a random location at the beginning of the simulation.
Within the simulation, I will take several measurements for each strategy and number of searchers (1 to 5), so that the number is statistically significant and use, for example, the means to compare which strategy is the most appropriate.

The model will be able to simulate several search strategies

*one step forward and then turn of random degree (-45 to 45 degrees), so random walk
*walk straight until it hits the edge of the forest or tree, then turn and continue walking straight
*first walk to the nearest corner of the forest and then a some kind of serpentine search
*possibly other strategies

'''Goals:'''

Finding the most appropriate strategy for finding a person in the forest depending on the number of people searching.

'''Agents:'''

*Searchers (e.g. police officers)
*Lost person

'''Parameters:'''

*Number of searchers
*Type of strategy
*Ticks needed to find person

'''Possible extensions:'''

*Searchers with certain pace of walking
*Finding the person won’t mean be at same location but seeing it for some distance (again certain ability of the searcher to see for certain distance)
*Cooperation of finders (formations, place distribution)
*Lost person will be moving when being looked for