Advanced Software Development for Complex Systems Modelling
Lead Research Organisation:
University of Sheffield
Department Name: Computer Science
Abstract
FLAME (Flexible Large-scale Agent Modelling Environment) is a piece of software which facilitates the simulation of many simple individuals (agents). It provides modelling capabilities for these agents, allowing variables and interaction rules to be set. Once the agent models are defined, simulations may be run, with emergent behaviour arising due to the interaction between agents over numerous time steps. This happens as each time step updates the agents, and their variables as necessary. Some examples of systems which may be simulated in such a manner can be found on the FLAME website (http://flame.ac.uk/projects/) and include both biological and economic systems.
My proposed research topic (Advanced Software Development for Complex Systems Modelling) will focus on the further development of the core methods and algorithms used within FLAME-II, a framework based on FLAME, to parallelise agent based applications. The research questions and hypotheses to be covered in this project are yet to be decided, though they will follow along the lines of "Is it possible to implement or improve functionality within FLAME-II allowing for efficient and viable simulation of ... using the FLAME-II framework?". The project title is broad enough to allow me to refine this question to a specific topic within a specific discipline over the first year or so of the project. Once the question(s) are decided upon I would allot another year plus leeway to implement the functionality to the required standard before using the remaining time to finalise a thesis and ensure the functionality is correct.
As my current area of interest with relation to the project is the simulation of biological systems, this is where my research would initially be directed at. This would likely involve individual and guided research into areas which may benefit from simulation as well as contact with experts to ensure that this is a correct assumption in their area. During the design and near the end of the implementation phase this input from experts would provide the necessary direction to create functionality of real use.
I would expect that the end result of the project would provide a specific topic an efficient, powerful potential for performing simulations and giving researchers within the area a new tool with which they may visualise systems, providing insight into the intricacies involved with the interaction of many distinct entities.
Resources
FLAME website - http://flame.ac.uk/
Brief overview of FLAME - http://flame.ac.uk/docs/overview.html
Dr Paul Richmond's related project, FLAME-GPU - http://www.flamegpu.com/
Agent-based modelling - http://www.pnas.org/content/99/suppl_3/7280.full
My proposed research topic (Advanced Software Development for Complex Systems Modelling) will focus on the further development of the core methods and algorithms used within FLAME-II, a framework based on FLAME, to parallelise agent based applications. The research questions and hypotheses to be covered in this project are yet to be decided, though they will follow along the lines of "Is it possible to implement or improve functionality within FLAME-II allowing for efficient and viable simulation of ... using the FLAME-II framework?". The project title is broad enough to allow me to refine this question to a specific topic within a specific discipline over the first year or so of the project. Once the question(s) are decided upon I would allot another year plus leeway to implement the functionality to the required standard before using the remaining time to finalise a thesis and ensure the functionality is correct.
As my current area of interest with relation to the project is the simulation of biological systems, this is where my research would initially be directed at. This would likely involve individual and guided research into areas which may benefit from simulation as well as contact with experts to ensure that this is a correct assumption in their area. During the design and near the end of the implementation phase this input from experts would provide the necessary direction to create functionality of real use.
I would expect that the end result of the project would provide a specific topic an efficient, powerful potential for performing simulations and giving researchers within the area a new tool with which they may visualise systems, providing insight into the intricacies involved with the interaction of many distinct entities.
Resources
FLAME website - http://flame.ac.uk/
Brief overview of FLAME - http://flame.ac.uk/docs/overview.html
Dr Paul Richmond's related project, FLAME-GPU - http://www.flamegpu.com/
Agent-based modelling - http://www.pnas.org/content/99/suppl_3/7280.full
Organisations
People |
ORCID iD |
| Paul Richmond (Primary Supervisor) | |
| James Pyle (Student) |
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/N509735/1 | 01/10/2016 | 30/09/2021 | |||
| 2112851 | Studentship | EP/N509735/1 | 01/10/2017 | 10/07/2022 | James Pyle |
| EP/R513313/1 | 01/10/2018 | 30/09/2023 | |||
| 2112851 | Studentship | EP/R513313/1 | 01/10/2017 | 10/07/2022 | James Pyle |