Temporal meta-optimiser based sensitivity analysis (TMSA) for agent-based models and applications in children’s services

Autor: Luke White, Shadi Basurra, Abdulrahman A. Alsewari, Faisal Saeed, Sudhamshu Mohan Addanki
Jazyk: angličtina
Rok vydání: 2024
Předmět:
Zdroj: Scientific Reports, Vol 14, Iss 1, Pp 1-16 (2024)
Druh dokumentu: article
ISSN: 2045-2322
DOI: 10.1038/s41598-024-59743-8
Popis: Abstract With current and predicted economic pressures within English Children’s Services in the UK, there is a growing discourse around the development of methods of analysis using existing data to make more effective interventions and policy decisions. Agent-Based modelling shows promise in aiding in this, with limitations that require novel methods to overcome. This can include challenges in managing model complexity, transparency, and validation; which may deter analysts from implementing such Agent-Based simulations. Children’s Services specifically can gain from the expansion of modelling techniques available to them. Sensitivity analysis is a common step when analysing models that currently has methods with limitations regarding Agent-Based Models. This paper outlines an improved method of conducting Sensitivity Analysis to enable better utilisation of Agent-Based models (ABMs) within Children’s Services. By using machine learning based regression in conjunction with the Nomadic Peoples Optimiser (NPO) a method of conducting sensitivity analysis tailored for ABMs is achieved. This paper demonstrates the effectiveness of the approach by drawing comparisons with common existing methods of sensitivity analysis, followed by a demonstration of an improved ABM design in the target use case.
Databáze: Directory of Open Access Journals
Nepřihlášeným uživatelům se plný text nezobrazuje