The Next Frontier in HTA Reviewing the Integration of Artificial Intelligence and Discrete Event Simulation for Resource Modeling
Author(s)
Syed Salleh, PhD1, Nishtha Neeraj, MSc2, Aishee Ghatak, MSc2, Tushar Srivastava, MSc1.
1ConnectHEOR, London, United Kingdom, 2ConnectHEOR, Delhi, India.
1ConnectHEOR, London, United Kingdom, 2ConnectHEOR, Delhi, India.
OBJECTIVES: This study explores how integrating artificial intelligence (AI), particularly machine learning, with discrete-event simulation (DES) can enhance resource modelling (RM) in healthcare. It examines the potential of AI-DES to capture real-world constraints and their impact on performance, costs, and outcomes, highlighting its relevance for health technology assessment (HTA) and improving the realism and policy value of economic evaluations.
METHODS: A two-level targeted literature review was conducted in PubMed (June-2020-June-2025). In level 1, we searched for studies integrating AI with DES specifically within the context of HTA. The scope was further expanded to level 2 to include AI-DES applications in broader healthcare operational research (OR), with a particular focus on RM involving capacity constraints (e.g., staff, beds) and their influence on system performance and patient outcomes.
RESULTS: The initial search yielded no AI-DES studies in HTA. Of 59 studies screened in level 2, 13 were reviewed and 9 included for data extraction. All focused on applying AI-DES models to resource management in healthcare. Common AI techniques included clustering and predictive modeling, often used to support scheduling, patient prioritization, and workflow optimization. Resources were defined as either single-use (e.g., test kits) or reusable (e.g., beds, staff). One study combined clustering with a stochastic, multi-objective DES to optimize chemotherapy scheduling under capacity constraints. AI enhanced DES by simulating bottlenecks, delays, and dynamic resource use. Most studies addressed capacity and throughput constraints using fixed or time-varying parameters.
CONCLUSIONS: AI-DES models show promise for optimizing healthcare operations but remain underexplored in HTA. While sometimes characterized as ‘black box’ tools, these models can enhance DES scalability and responsiveness. Gaps in validation and real-world use of AI-DES in healthcare setting highlight the need for future research to focus on how these models can be leveraged to support economic evaluations and inform decision-making within HTA frameworks.
METHODS: A two-level targeted literature review was conducted in PubMed (June-2020-June-2025). In level 1, we searched for studies integrating AI with DES specifically within the context of HTA. The scope was further expanded to level 2 to include AI-DES applications in broader healthcare operational research (OR), with a particular focus on RM involving capacity constraints (e.g., staff, beds) and their influence on system performance and patient outcomes.
RESULTS: The initial search yielded no AI-DES studies in HTA. Of 59 studies screened in level 2, 13 were reviewed and 9 included for data extraction. All focused on applying AI-DES models to resource management in healthcare. Common AI techniques included clustering and predictive modeling, often used to support scheduling, patient prioritization, and workflow optimization. Resources were defined as either single-use (e.g., test kits) or reusable (e.g., beds, staff). One study combined clustering with a stochastic, multi-objective DES to optimize chemotherapy scheduling under capacity constraints. AI enhanced DES by simulating bottlenecks, delays, and dynamic resource use. Most studies addressed capacity and throughput constraints using fixed or time-varying parameters.
CONCLUSIONS: AI-DES models show promise for optimizing healthcare operations but remain underexplored in HTA. While sometimes characterized as ‘black box’ tools, these models can enhance DES scalability and responsiveness. Gaps in validation and real-world use of AI-DES in healthcare setting highlight the need for future research to focus on how these models can be leveraged to support economic evaluations and inform decision-making within HTA frameworks.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
HTA324
Topic
Health Technology Assessment
Topic Subcategory
Decision & Deliberative Processes
Disease
No Additional Disease & Conditions/Specialized Treatment Areas