Immediate treatment is of extreme importance for stroke patients. However, providing fast enough treatment for stroke patients is far from trivial, mainly due to logistical challenges and difficulties in diagnosing the correct stroke type. One way to reduce the time to treatment is to use so-called Mobile Stroke Units (MSUs), which allows to diagnose and provide treatment for stroke patients already at the patient scene. A well-designed stroke transport policy is vital to improve the access to treatment for stroke patients. Simulation and mathematical optimization are useful approaches for assessing and optimizing stroke transport policies, without endangering the health of the patients.

The main purpose of this thesis is to contribute to improving the situation for stroke patients and to reducing the social impacts of stroke. The aim is to study how to use simulation and optimization to achieve improved analysis and planning of prehospital stroke care. In particular, we focus on assessing the potential use of MSUs in a geographic area. In this thesis, optimization is used to identify the optimal locations of MSUs, and simulation is used to assess different stroke transport policies, including MSU locations. The results of this thesis aim to support public health authorities when making decisions in the prehospital stroke care domain.

In order to fulfill the aim of this thesis, we develop and analyze a number of different simulation and optimization models. First, we propose a macro-level simulation model, an average time to treatment estimation model, used to estimate the expected time to treatment for different parts of a geographic region. Using the proposed model, we generate two different MSU scenarios to explore the potential benefits of employing MSUs in Sweden’s southern healthcare region (SHR).  

Second, we present an optimization model to identify the best placement of MSUs while making a trade-off between the efficiency and equity perspectives, providing maximum population coverage and equal service for all patients, respectively. The trade-off function used in the model makes use of the concepts of weighted average time to treatment to model efficiency and the time difference between the expected time to treatment for different geographical areas to model equity. In a scenario study applied in the SHR, we evaluate our optimization model by comparing the current situation with three MSU scenarios, including 1, 2, and 3 MSUs.

Third, we present a micro-level discrete event simulation model to assess stroke transport policies, including MSUs, allowing us to model the behaviors of individual entities, such as patients and emergency vehicles, over time. We generate a synthetic set of stroke patients using a Poisson distribution, used as input in a scenario study.

Finally, we present a modeling framework with reusable components, which aims to facilitate the construction of discrete event simulation models in the emergency medical services domain. The framework consists of a number of generic activities, which can be used to represent healthcare chains modeled in the form of flowcharts. As the framework includes activities and policies modeled on the general level, the framework can be used to create models only by providing input data and a care chain specification. We evaluate the framework by using it to build a model for simulating EMS activities related to the complex case of acute stroke.

Rapid treatment is crucial for minimizing the consequences of a stroke. However, logistical challenges and the complexity of accurate stroke diagnosis often impede timely and effective treatment. One way to reduce time to treatment is the use of so-called mobile stroke units (MSUs), which are specialized ambulances equipped to diagnose and treat stroke patients on site. The adequate planning and optimization of prehospital stroke transport policies involving MSUs can help reduce delays in accessing treatment. Mathematical optimization and simulation are useful approaches for optimizing and assessing different stroke transport policies without endangering patient’s health.

The aim of this thesis is to explore how optimization and simulation can improve the analysis and planning of prehospital stroke care. Specifically, optimization is used to determine optimal MSU placements, while simulation is applied to evaluate stroke transport policies, including those involving MSUs. To achieve this aim, the thesis is structured around four main objectives, in which we develop and analyze a number of different optimization and simulation models. First, the MSU placement problem is solved using an exhaustive search algorithm and formulated as a mixed-integer linear programming model to determine optimal MSU placements. The objective of solving this problem is to make a trade-off between efficiency and equity, ensuring maximum population coverage and equitable service across a region. Second, macro-level and micro- level simulation models are proposed to evaluate various stroke transport policies, including MSUs. Third, a simulation modeling framework is introduced to enable the construction of discrete event simulation models for emergency medical services (EMS) policy analysis, supporting flexible and adaptive simulations of real-world EMS operations. The framework incorporates various decision policies, such as emergency vehicle selection, dispatch type (single and co-dispatch) selection, and hospital selection, allowing for the evaluation of stroke transport policies across different stroke types. Lastly, dynamic travel time calculations and machine learning-based travel time estimations are integrated into the framework to enhance the flexibility and reliability of EMS simulations.

Through scenario studies conducted in Sweden’s Southern Healthcare Region, this research demonstrates how optimization and simulation can support effective stroke transport policy planning and improve decision-making in prehospital stroke care. The identified MSU placements, along with the evaluated dispatch policies, highlight significant potential for reducing the time to diagnosis and treatment for different types of strokes. Faster time to treatment not only enhances overall stroke care delivery but also improves patient outcomes by reducing stroke-related disabilities. The findings underscore the value of these approaches in guiding EMS policy design, ultimately contributing to better patient outcomes and reduced social impacts of stroke. The results of this thesis aim to assist public health authorities in making informed decisions to optimize prehospital stroke care.

Rapid treatment is crucial for minimizing the consequences of a stroke. However, logistical challenges and the complexity of accurate stroke diagnosis often impede timely and effective treatment. One way to reduce time to treatment is the use of so-called mobile stroke units (MSUs), which are specialized ambulances equipped to diagnose and treat stroke patients on site. The adequate planning and optimization of prehospital stroke transport policies involving MSUs can help reduce delays in accessing treatment. Mathematical optimization and simulation are useful approaches for optimizing and assessing different stroke transport policies without endangering patient’s health. The aim of this thesis is to explore how optimization and simulation can improve the analysis and planning of prehospital stroke care. Specifically, optimization is used to determine optimal MSU placements, while simulation is applied to evaluate stroke transport policies, including those involving MSUs. To achieve this aim, the thesis is structured around four main objectives, in which we develop and analyze a number of different optimization and simulation models. First, the MSU placement problem is solved using an exhaustive search algorithm and formulated as a mixed-integer linear programming model to determine optimal MSU placements. The objective of solving this problem is to make a trade-off between efficiency and equity, ensuring maximum population coverage and equitable service across a region. Second, macro-level and micro- level simulation models are proposed to evaluate various stroke transport policies, including MSUs. Third, a simulation modeling framework is introduced to enable the construction of discrete event simulation models for emergency medical services (EMS) policy analysis, supporting flexible and adaptive simulations of real-world EMS operations. The framework incorporates various decision policies, such as emergency vehicle selection, dispatch type (single and co-dispatch) selection, and hospital selection, allowing for the evaluation of stroke transport policies across different stroke types. Lastly, dynamic travel time calculations and machine learning-based travel time estimations are integrated into the framework to enhance the flexibility and reliability of EMS simulations. Through scenario studies conducted in Sweden’s Southern Healthcare Region, this research demonstrates how optimization and simulation can support effective stroke transport policy planning and improve decision-making in prehospital stroke care. The identified MSU placements, along with the evaluated dispatch policies, highlight significant potential for reducing the time to diagnosis and treatment for different types of strokes. Faster time to treatment not only enhances overall stroke care delivery but also improves patient outcomes by reducing stroke-related disabilities. The findings underscore the value of these approaches in guiding EMS policy design, ultimately contributing to better patient outcomes and reduced social impacts of stroke. The results of this thesis aim to assist public health authorities in making informed decisions to optimize prehospital stroke care.