kth.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
An agent-based model approach to computational epidemiology
KTH, School of Electrical Engineering and Computer Science (EECS), Computer Science.
KTH, School of Electrical Engineering and Computer Science (EECS), Computer Science.
2022 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesisAlternative title
En tillämpning av en agentbaserad modell för beräkningsmässig epidemiologi (Swedish)
Abstract [en]

The aim of this work was to compare different Non-pharmaceutical interventions (NPIs) in a simulated environment modeling Sweden with 10 000 000 (ten million) agents. Simulating the performance of these NPIs can be beneficial to policymakers and other government officials in order to give the power to make the right decision in the face of an epidemic. In this work, an agent-based model (ABM) was used, rather than estimating an ordinary differential equation, which is often done in the field of computational epidemiology. An existing ABM was used to perform this task. We found that using various interventions could greatly decrease the number of infected agents as well as fatalities at the end of the simulations. Five different types of interventions were simulated, these were; baseline, lockdown, home quarantine, case isolation and mask usage. The results from the interventions were compared against a baseline of no intervention. The most effective intervention that was simulated by far was total lockdown, reducing infected agents by 99.4% and fatalities by 99.6% from baseline. In second place, home quarantine reduced the number of infected agents by 38.0% and fatalities by 45.1%. We found that the reduction in infected agents was much greater in the simulated environment than in the real world, when compared with the results from another study on humans rather than simulated agents. There was a certain similarity between how well the methods ranked in effectiveness in the simulation and in reality. However, they were not equal. The results and methods of this work are probably not suitable as a proxy for estimating real-world effectiveness of NPIs, but may be used to compare the different NPIs against each other.

Abstract [sv]

Detta arbete ämnar till att jämföra olika icke-farmaceutiska insatser (NPIs) i en simulerad miljö som avser att modellera Sverige med 10 000 000 (tio miljoner) invånare. Att simulera prestandan av dessa insatser kan vara gynnsamt för olika sorters beslutsfattare i ett lands styre för att ge dem makten att ta rätt beslut under en epidemi. I detta arbete användes en agentbaserad modell (ABM), snarare än att estimera en ordinär differentialekvation som man ofta gör inom beräkningsepidemiologi. En existerande ABM användes för att utföra denna uppgift. Vi fann att olika insatser kunde ha stor inverkan på antalet påverkade agenter, samt antalet dödsfall vid slutet av simuleringarna. Fem olika typer av insatser var simulerade, dessa var; basfallet, nedstängning av samhället, hemkarantän och maskanvändning. Resultaten från insatserna jämfördes mot ett standardvärde med ingen insats alls. Den överlägset mest effektiva insatsen var en nedstängning av samhället (lockdown), som minskade antalet påverkade agenter med 99,4% och antalet dödsfall med 99,6% från standardvärde. På andra plats kom hemkarantän som minskade antalet påverkade agenter med 38,0% och antalet dödsfall med 45,1%. Vi fann att minskningen av både antalet påverkade agenter samt dödsfall var mycket större i den simulerade domänen än i verkligheten, när vi jämförde de simulerade resultaten med resultaten från en studie på människor snarare än simulerade agenter. Det fanns en viss likhet i hur bra metoderna rankade mot varandra i effektivitet i simuleringen och verkligheten, men likheten var inte fullständig. Resultaten och metoderna från denna rapport är sannolikt inte en bra proxy för att estimera hur dessa insatser skulle prestera i verkligheten, men kan möjligtvis användas för att jämföra de olika insatserna mot varandra.

Place, publisher, year, edition, pages
2022. , p. 30
Series
TRITA-EECS-EX ; 2022:459
National Category
Computer and Information Sciences
Identifiers
URN: urn:nbn:se:kth:diva-319681OAI: oai:DiVA.org:kth-319681DiVA, id: diva2:1701462
Subject / course
Computer Science
Educational program
Master of Science in Engineering - Computer Science and Technology
Supervisors
Examiners
Available from: 2022-10-07 Created: 2022-10-06 Last updated: 2022-10-07Bibliographically approved

Open Access in DiVA

fulltext(711 kB)133 downloads
File information
File name FULLTEXT01.pdfFile size 711 kBChecksum SHA-512
2cbaef066533047f08e13e1db21a4de6258c17cf26bb5cd9b1cab47e5b7b63eb642ae86f42f8d6a18986a613cfeebc733e7a25d69a5b08fc4afec2cdd65d54cb
Type fulltextMimetype application/pdf

By organisation
Computer Science
Computer and Information Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 134 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 217 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf