28/10/2024, 10:56 Agent based modelling
Agent based modelling
Due 13 Nov 2023 by 15:00
Points 100
Submitting a file upload
File types nlogo
Available 9 Oct 2023 at 0:00 - 20 Nov 2023 at 23:59
This assignment was locked 20 Nov 2023 at 23:59.
Notes for students
Regulations governing assessment offences including Plagiarism and Collusion are available from
https://www.herts.ac.uk/__data/assets/pdf_file/0007/237625/AS14-Apx3-Academic-
Misconduct.pdf (https://www.herts.ac.uk/__data/assets/pdf_file/0007/237625/AS14-Apx3-
Academic-Misconduct.pdf) (UPR AS14).
Guidance on avoiding plagiarism can be found here:
https://herts.instructure.com/courses/61421 (https://herts.instructure.com/courses/61421) (see
the Referencing section)
a score of 50% or above represents a pass mark.
late submission of any item of coursework for each day or part thereof (or for hard copy
submission only, working day or part thereof) for up to five days after the published deadline,
coursework relating to modules at Level 7 submitted late (including deferred coursework, but with
the exception of referred coursework), will have the numeric grade reduced by 10 grade points
until or unless the numeric grade reaches or is 50. Where the numeric grade awarded for the
assessment is less than 50, no lateness penalty will be applied.
This Assignment assesses the following module Learning Outcomes (from Definitive Module
Document):
1. Demonstrate knowledge and understanding of a variety of Artificial Life techniques and methods
applicable across domains including computer science and robotics.
2. Demonstrate the ability to critically evaluate Artificial Life systems.
3. Demonstrate the ability to implement an Artificial Life system according to a defined specification.
Agent Based Modelling
In this project you will be developing a model to ascertain important factors the impact the spread of
an infectious disease among a population of agents within a particular timeframe. You will have 2
populations of agents that are infected with a virus which will spread among the populations. You will
analyse factors that impact the spread and mortality rate such as population density, social
distancing, self-isolation and the amount of time the virus can go undetected whist transmissible.
https://herts.instructure.com/courses/107328/assignments/246810 1/9
28/10/2024, 10:56 Agent based modelling
Recording overview:
0:00 / 38:25
Assignment Brief:
In this project you must develop a model in order to ascertain the spread of an infectious disease
among a population of agents within a particular timeframe. The purpose of this model is to analyse
the following aspects of a disease:
Assess the impact of agents staying local within a designated region vs free movement
Assess the impact of social distancing between agents vs not social distancing
Assess the impact of individuals who are infected self-isolating vs not self-isolating
https://herts.instructure.com/courses/107328/assignments/246810 2/9
28/10/2024, 10:56 Agent based modelling
Since this is meant to provide a relatively realistic representation of a real world situation, the criteria
for this model is clearly outlined below.
You must design simulation in NetLogo that adheres to the following requirements:
1. Overall Model Requirements (5 marks)
Your model MUST run without error. If your model crashes you will get no marks.
The following code must be placed at the beginning of your model (copy and paste this in):
extensions [csv]
globals[
;Variables to be used for tutor marking
student_id
student_name
student_score
student_feedback
;Variables for your analysis
most_effective_measure
least_effective_measure
population_most_affected
population_most_immune
self_isolation_link
population_density
total_infected_percentage
cyan_infected_percentage
lime_infected_percentage
total_deaths
cyan_deaths
lime_deaths
total_antibodies_percentage
cyan_antibodies_percentage
lime_antibodies_percentage
stored_settings
]
turtles-own[
infected_time
antibodies
group
]
https://herts.instructure.com/courses/107328/assignments/246810 3/9
28/10/2024, 10:56 Agent based modelling
Your submitted file must not use any of the following commands:
clear-all
clear-globals
clear-ticks
clear-drawing
clear-all-plots
clear-output
stop
show
print
write
If any of these commands are found in your model this will result in a 5 mark deduction per command
found.
The default turtles must be used to represent the agents in the model, no additional breeds can
be created.
You must not use the global variables student_score or student_feedback at all in your model, if
this is found in your model this will result in a 5 mark deduction.
1. The world must be 101 patches wide by 101 patches high.
The origin of the world must be at the centre (center).
The world must wrap both horizontally and vertically.
2. The patch size must be set to 5.
3. The framerate must be set to 20 frames per second.
1. World setup (5 marks)
The model must have a function called setup_world that initiates the following actions in your model
when called:
The tick counter must be reset
All turtles must be cleared from the world
1. The world must be split into 2 distinct halves which are both of an (almost) equal size that are one
continuous body of color (see example image). One half must be identified with the patches being
set to the color lime whilst the other half must be set to the color cyan.
The global variables student_id and student_name must be set to your student ID number and full
name respectively.
The model must contain the following global variables (these can be defined in the code or using
the interface) and be set to the following values when submitted:
1. cyan_population = 1000
2. lime_population = 500
https://herts.instructure.com/courses/107328/assignments/246810 4/9
28/10/2024, 10:56 Agent based modelling
3. initially_infected = 20
4. infection_rate = 15
5. survival_rate = 70
6. immunity_duration = 250
7. undetected_period = 70
8. illness_duration = 300
9. travel_restrictions = false
10. social_distancing = false
11. self_isolation = false
The setup_world function must not call the initialise_agent function or the run_model function.
1. Agents setup (10 marks)
The model must have a function called setup_agents that initiates the following actions in your model
when called:
Cyan population
Create a population of turtles that are set to the color grey, size 2, and are randomly placed on the
area of cyan patches. The number of turtles created must be determined by the variable
cyan_population. All of these agents must be initialised with their own antibodies variable set to 0
and their group to “cyan turtle”. NOTE: antibodies must always be an integer value.
Lime population
Create a population of turtles that are set to the color grey, size 2, and are randomly placed on the
area of lime patches. The number of turtles created must be determined by the variable
lime_population. All of these agents must be initialised with their own antibodies variable set to 0
and their group to “lime turtle”.
Initially infected individuals
Create a number of infected turtles in each population (cyan and lime) determined by the global
variable initially_infected. For example if initially_infected is set to 20 there would be 20 individuals
in the cyan population infected and 20 individuals in the lime population infected.
The setup_agents function must not call the setup_world function or the run_model function.
1. Running requirements (50 marks)
The model must have a function called run_model that initiates the following actions in your model
when called:
https://herts.instructure.com/courses/107328/assignments/246810 5/9
28/10/2024, 10:56 Agent based modelling
The tick command must be called to add 1 to the tick counter.
Turtle Behaviours
The turtles in your model must behave according to the following parameters:
The speed at which turtles move must be set to 0.2 per step taken and they must always move in
a forwards direction (relative to their heading).
The orientation (heading) of turtles must be set randomly when wandering but must be limited to a
range of 40 degrees (20 to the left and 20 to the right) per step taken, the only exception to this is
when avoiding other agents, in this instance a maximum turn of 90 degrees is permitted.
Turtles must follow the rules of the global variables travel_restrictions, social_distancing and
self_isolation as follows:
Travel Restrictions
If the travel_restrictions variable is set to true the turtle must stay within its own region (lime for lime
turtles, cyan for cyan turtles), if the turtle is not within its own region (i.e. the variable was changed
mid-simulation) the turtle must take the shortest route back to its own region following any other
conditions instated (i.e. social_distancing or self_isolation).
Social distancing
If the social_distancing variable is set to true the turtle must keep a minimum distance of 1 patch
between itself and another agent when moving (i.e. check to see if another agent is in front before
moving forward, a strategy to avoid collisions).
Self-isolation
If the self_isolation variable is set to true the agent must stop moving and turn orange to indicate it is
in self-isolation when infected after the limit undetected_period (i.e. the illness_duration -
undetected_period).
Agents become infected when they are within a radius of 1 of another agent that is infected that is
not in self-isolation and does not have antibodies.
The probability of agents becoming infected by coming into contact with other infected agents
must be determined by the global variable infection_rate which must range from 0 to 100 (0 = 0%
chance of becoming infected, 100 = 100% chance of becoming infected).
Agents who are infected must turn red to indicate this and also set their infected_time variable
(stored in the agents own variable) to the value stated in the global variable illness_duration and
reduce by 1 per tick to indicate the passing of time of the illness.
The survival rate of an agent must be dictated by the variable survival_rate (i.e. if survival_rate is
set to 80 this would mean that 80% of the time agents survive the illness)
If an agent survives an infection at the end of the infected_time the agent must develop antibodies
(stored in the agents own variable called antibodies) which must be set according to the global
variable immunity_duration and reduce by 1 per tick to indicate the reduction of antibodies over
time (again a count down timer). An agent that has antibodies must turn black while they have
antibodies as a visual representation.
https://herts.instructure.com/courses/107328/assignments/246810 6/9
28/10/2024, 10:56 Agent based modelling
Model outputs
Whilst run_model is being called (whist your model is running) you must update the following global
variables in real-time (after each tick):
The global variable total_infected_percentage must show the percentage of the current total
living population that are infected.
The global variable cyan_infected_percentage must show the percentage of the current cyan
living population that are infected.
The global variable lime_infected_percentage must show the percentage of the current lime
living population that are infected.
The global variable total_deaths must be updated on every tick and show the number of turtles
who have died in total.
The global variable cyan_deaths must be updated on every tick and show the number of cyan
turtles who have died in total.
The global variable lime_deaths must be updated on every tick and show the number of lime
turtles who have died in total.
The global variable total_antibodies_percentage must show the percentage of the current total
population that are have antibodies and are immune to infection.
The global variable cyan_antibodies_percentage must show the percentage of the current cyan
population that are have antibodies and are immune to infection.
The global variable lime_antibodies_percentage must show the percentage of the current lime
population that are have antibodies and are immune to infection.
1. Your analysis (30 marks)
By the end of your implementation you should have a working model that you can use to perform
some analysis. You must have a function called my_analysis that populate a series of global variables
with your answers. This must be a self-contained function and not call any other function in your
model.
5.1. Most effective measure
Having run the model with the default population over 5,000 ticks in different conditions, what do you
think the single most effective measure was for reducing the spread of the virus?
1 = social distancing
2 = self-isolation
3 = stay local
Set the global variable most_effective_measure to one of the 3 answers above (just set the variable
to the number).
https://herts.instructure.com/courses/107328/assignments/246810 7/9
28/10/2024, 10:56 Agent based modelling
Least effective measure
Having run the model with the default population over 5,000 ticks in different conditions, what do you
think the single least effective measure was for reducing the spread of the virus?
1 = social distancing
2 = self-isolation
3 = stay local
Set the global variable least_effective_measure to one of the 3 answers above (just set the variable
to the number).
Population with highest mortality rate
Having run the model with the default populations which population had the highest mortality rate
(total number of deaths)?
1 = cyan
2 = lime
3 = equally affected
Set the global variable population_with_highest_mortality_rate to one of the 3 answers above (just
set the variable to the number).
Population with highest immunity
Having run the model with the default populations which population gains greater immunity?
1 = cyan
2 = lime
3 = equally affected
Set the global variable population_most_immune to one of the 3 answers above (just set the variable
to the number).
Links with self-isolation
Which global variable has the most impact on the self-isolation measure?
1 = cyan_population
2 = lime_population
3 = initially_infected
https://herts.instructure.com/courses/107328/assignments/246810 8/9
28/10/2024, 10:56 Agent based modelling
4 = infection_rate
5 = survival_rate
6 = immunity_duration
7 = undetected_period
8 = illness_duration
Set the global variable self_isolation_link to one of the 8 answers above (just set the variable to the
number).
Impact of population density
Running the model with all of the default settings but having the travel_restrictions switched on what
happens to the populations after 20,000 ticks?
1 = both populations reduce equally
2 = the lime population is completely eliminated
3 = the cyan population is completely eliminated
4 = the virus dies out
5 = both populations settle to a similar density and the virus dies out
6 = both populations settle to a similar density
Set the global variable population_density to one of the 6 answers above (just set the variable to the
number).
What to Submit
This assignment requires the submission of 1 NetLogo file version 6.1.0 or above. The file should be
named as follows: <your student id>_ABM.nlogo
https://herts.instructure.com/courses/107328/assignments/246810 9/9
