System Dynamics Model Research Articles

Using routine health care data to develop and validate a system dynamics simulation model of frailty trajectories in an ageing population

ABSTRACT Frailty is common in older adults and has a substantial impact on patient outcomes and service use. Information to support service planning, including prevalence in middle-aged adults and patterns of frailty progression at population level, is scarce. This paper presents a system dynamics model describing the dynamics of frailty and ageing within a population of patients aged ≥50, based on linked data for 2.2 million patients from primary care practices in England. The purpose of the model is to estimate the incidence and prevalence of frailty in an ageing population over time. The model was developed in consultation with stakeholders (patients, carers, clinicians, and commissioners) and validated against another large dataset (1.38 million patients) from Wales. It was then scaled up to the population of England, using Office for National Statistics projections (to 2027). The baseline results, subject to the assumption that the frailty transition parameters remain constant over this period, suggest that the number of people living with frailty will increase as the population ages, and that those with mild-moderate frailty are likely to have the greatest impact on demand. This paper focuses on model development and validation, highlighting the benefits and challenges of using large routine health datasets.

Simulation and applicability evaluation of providing different grades of water based on household functional needs in urban areas.

Providing different grades of water is a promising solution to address the challenges of urban water supply, including water quality, quantity, and energy consumption. However, quantifying the effectiveness of this strategy and understanding its economic, environmental, and social impacts remain significant challenges. This study introduces a simulation-based method to predict household water use and evaluate the applicability and sustainability of the water supply system. A system dynamics model was developed to simulate household water behavior under different water supply scenarios, incorporating different water quality combinations, price mechanisms, and regional population composition. The results were analyzed regarding the optimal water completion degree, user satisfaction, water supply yield, and water volume stability to evaluate the system's applicability. Our findings suggest that scenarios involving simultaneous provision of high-quality and common water are deficient due to overlapping functions. In addition, a dynamic water pricing mechanism can improve user satisfaction and supply stability, especially in the area with a large population. Concurrent supply of high-quality, low-quality, and reclaimed water emerges as an efficient alternative, achieving dual objectives of resource conservation and water quality improvement.

Spatiotemporal simulation of blue-green space pattern evolution and carbon storage under different SSP-RCP scenarios in Wuhan

Rapid socioeconomic growth has altered land use patterns, resulting in a surge in worldwide CO2 emissions, triggering global climate challenges and adversely affecting human health, safety, and sustainable socioeconomic development. As a result, immediate action is required to undertake climate mitigation and adaptation strategies. This study, based on the causal logic of climate change, blue-green space patterns, and carbon emissions, uses the system dynamics (SD) model, patch-generating land use simulation (PLUS) model, and integrated valuation of ecosystem service and trade-offs (InVEST) models to simulate the evolution of blue-green space patterns and predict the spatial distribution of carbon storage in Wuhan to 2060 from 2030 under three SSP-RCP scenarios from CMIP6 and investigates their mechanisms. The findings show that across various SSP-RCP scenarios, the blue-green space patterns in Wuhan would decline over the next 30 years, with green spaces decreasing to some amount and blue spaces growing marginally. The carbon storage is also expected to decline due to the shrinking blue-green space patterns. The SSP126 scenario has the least shrinkage of blue-green spaces, resulting in a reduction of 7.18Tg in carbon storage. Under the SSP245 scenario, the expansion of non-blue-green spaces encroaches on blue-green spaces, resulting in an 8.13 Tg decrease in carbon storage. Across the SSP585 scenario, non-blue-green spaces expand the fastest, resulting in the highest loss of blue-green spaces and a considerable drop in carbon storage of 11.67 Tg. This research is extremely important for optimizing regional land use patterns, coordinating green and high-quality development in Wuhan, and assisting with the implementation of urban climate change adaptation plans.

The dynamic model of Iran’s electrical energy supply system based on water-food-energy-climate change nexus

PurposeEnsuring energy security and controlling the share of energy in climate change are the two main challenges of the energy sector in the path of a sustainable future. This study has designed the dynamic model of Iran’s electrical energy supply system based on the water-food-energy-climate change nexus in order to identify sustainable policies for the supply of electrical energy resources and adaptation to Iran’s climate change process.Design/methodology/approachFor this purpose, first, the system dynamics model was designed with the participation of policymakers and the supply and demand data of water-food-energy resources and the trend of climate change and economic growth in Iran. After validation, the model is simulated in the 30-year horizon (2020–2050), and according to the results of the Monte Carlo sensitivity analysis, Iran’s electric energy supply policies are based on four strategies, including (1) Electric energy supply based on electric energy supply management from non-renewable sources, (2) Development of electrical energy supply based on management of energy supply from renewable sources, (3) Electrical energy supply based on electrical energy demand management and (4) Electrical energy supply based on adaptation to climate changes. By identifying and applying the policies of each strategy separately, the model was tested and the results were compared.FindingsBased on the implementation of the combination of selected policies in the model, the following policies have been proposed: 16% development of nuclear power plants, 18% reduction in the ratio of production of gas power plants to the total production of non-renewable power plants, and an increase in the production of combined cycle power plants through conversion of gas to combined cycle, energy aggregation and development of heat recovery systems in industrial units by 32%, a decrease of 5% Energy transmission and distribution losses, per capita reduction of energy consumption from 0.926 to the global average of 0.182 (MW) per year, management of water demand in the food sector by increasing irrigation efficiency to about 85%, a 27% increase in the area of land under the irrigation network, and reducing losses Food amounting to the global average of 0.9 m tons per year.Originality/valueThe proposed model is an application of system dynamics in the field of policymaking to ensure the security of electrical energy resources, taking into account the water-food-energy-climate changes nexus. The model is a valuable tool for policymakers in planning the sustainable management of resources in the path of adapting to climate change.

DESTiny, an online farm-wide tool to estimate the net carbon emissions of a pasture-based dairy farms in South Africa

The dairy farming industry is an important contributor to both South Africa’s economy and food security. However, dairy in South Africa, like elsewhere in the world, has come under scrutiny because of animal welfare concerns and contributions to greenhouse gas (GHG) emissions. To address GHG mitigation, we have constructed a farm-level system dynamics model to assess both the emissions, and capture and storage of carbon (C) on farms, to determine whether farms are net emitters of C (i.e., sources) or sequestrators (i.e., sinks). We have considered nutrient flows, the type and volume of feed, the effluent management system, various parameters relating to the herd dynamics, and the overall effect on the farm economy. The resulting online Dairy Environment Sustainability Tool (DESTiny) can help dairy farmers adopt sustainable practices, as well as improve competitiveness and financial sustainability while reducing the farm’s emissions profile, thereby building value-chain and consumer trust. DESTiny can be considered a science-informed evidence-based tool for estimating, monitoring and understanding nutrient and C flows in dairy production systems. It is also a web-based tool (see https://assetresearch.org.za/destiny-tool/) with a user-interface which allows remote users, researchers, practitioners, farmers and technicians ease of access while integrating the system dynamics models with on-farm realities.

Urban pandemic governance personal protective equipment allocation strategies: a system dynamics simulation

In the early days of the urban pandemic, many cities had personal protective equipment (PPE) shortages, which adversely affected urban pandemic governance. Using the COVID-19 strategies employed in Wuhan as the pivotal case study, this study sought to determine effective strategies to optimize city PPE distribution. System dynamics modeling was employed to explore the influence of PPE allocation strategies on pandemic control measures. It was found that the most effective method for controlling a pandemic was to supply PPE in a specific order: medical staff, patients, and out-of-home citizens. Further, prioritizing universal PPE access over adhering to recommended replacement frequencies was found to be more effective in protecting public health. These findings offer vital insights for policy formulation and pandemic preparedness planning to reduce infection rates and fatalities.

Dynamic Simulation and Reduction Path of Carbon Emission in “Three-Zone Space”: A Case Study of a Rapidly Urbanizing City

Understanding the current and future net carbon emission trajectories in “Three-Zone Space” is crucial for China to promote the formation of a low-carbon development pattern in territorial space and realize carbon neutrality. Taking Wuhan as the study area, we developed carbon emission and sequestration inventories for “Three-Zone Space”. Key driving factors of net carbon emissions were analyzed using the logarithmic mean division index, and future emissions and sequestration under six scenarios were projected with a system dynamics model. The optimal emission reduction pathway was identified through the intelligent decision-making index analysis. Our results show that Wuhan’s net carbon emission increased from 18.589 Mt in 2000 to 42.794 Mt in 2020. The emissions during this period primarily came from urban production space and urban living space. Economic development is the primary factor contributing to the increase in net carbon emissions (36.412 Mt). The efficiency of territorial space utilization is the strongest mitigator of net carbon emissions, reducing net carbon emissions by 74.341 Mt (accounting for 42.06% of total emissions). The comprehensive scenario is the most effective for net carbon emission reduction in urban and ecological spaces, while the technological progress scenario provides the greatest reduction potential in agricultural spaces. These findings provide actionable insights for optimizing spatial planning, enhancing ecological restoration, and adopting low-carbon agricultural technologies to achieve targeted emissions reductions in “Three-Zone Space”. The results of this study can further provide scientific basis for the formulation of targeted emission reduction measures for “Three-Zone Space” and guide the construction of low-carbon territorial space patterns.

Can system dynamics explain long-term hydrological behaviors? The role of endogenous linking structure

Abstract. Hydrological models with conceptual tipping bucket and process-based evapotranspiration formulations are the most common tools in hydrology. However, these models consistently fail to replicate long-term and slow dynamics of a hydrological system, indicating the need for model augmentation and a shift in formulation approach. This study employed an entirely different approach – system dynamics – towards more realistic replication of the observed slow hydrological behaviors at inter-annual and inter-decadal scales. Using the headwaters of Baiyang Lake in China as a case study, the endogenous linking structure of the hydrological system was gradually unraveled from 1982 to 2015 through wavelet analysis, Granger's causality test, and a system dynamics model. The wavelet analysis and Granger's causality test identified a negatively correlated and bidirectional causal relationship between actual evapotranspiration and catchment water storage change across distinct climatic periodicities, and the system dynamics approach suggested a combined structure of a vegetation reinforcing feedback and a soil water–vegetation balancing feedback in the hydrological system. The system dynamics' structure successfully captured the slow hydrological behaviors under both natural and human-intervention scenarios, demonstrating a self-sustained oscillation arising within the system's boundary. Our results showed that the interaction between the vegetation structure and the soil-bound water dominates the hydrological process at an inter-annual scale, while the interaction between the climatic oscillation and the soil-water-holding capacity dominates the hydrological process at an inter-decadal scale. Conventional hydrological models, which typically employ physiological-based evapotranspiration formulations and assume invariable soil characteristics, ignore vegetation structure change at the inter-annual scale and soil-water-holding capacity change at the inter-decadal scale, leading to failure in predicting the observed long-term hydrological behaviors. The system dynamics model is in its early stage with applications primarily confined to water-stressed regions and long-term scales. However, the novel insights proposed in our study, including the different hierarchies corresponding to distinct mechanisms and timescales and the endogenous linking structure among stocks being a more important driver of the hydrological behaviors, offer potential solutions for better understanding a hydrological system and guidelines for improving the configuration and performance of conventional hydrological models.

A Review of Transportation 5.0: Advancing Sustainable Mobility Through Intelligent Technology and Renewable Energy

Transportation 5.0 is an advanced and sophisticated system combining technologies with a focus on human-centered design and inclusivity. Its various components integrate intelligent infrastructure, autonomous vehicles, shared mobility services, green energy solutions, and data-driven systems to create an efficient and sustainable transportation network to tackle modern urban challenges. However, this evolution of transportation is also intended to improve accessibility by creating environmentally benign substitutes for traditional fuel-based mobility solutions, even when addressing traffic management and control issues. Consequently, to promote synergy for sustainability, the diversified nature of the Transportation 5.0 components ought to be efficiently and effectively managed. Thus, this study aims to reveal the involvement of Transportation 5.0 core component prediction in the sustainable transportation system through a systematic literature review. This study also contemplates the causal model under system dynamics modeling in order to address sustainable solutions and the movement toward sustainability in the context of Transportation 5.0. From this review, in addition to the developed causal model, it is identified that every core component management method in the sustainable Transportation 5.0 system reduces environmental impact while increasing passenger convenience and the overall efficiency and accessibility of the transport network, with greater improvements for developing nations. As the variety of transportation options, including electric vehicles, is successfully integrated, this evolution will eventually enable shared mobility, green infrastructure, and multimodal transit options.

Once concessioned, twice shy? Asymmetries of “reconcession” in national PPP programs

In Public-Private Partnerships, theoretically, once the concession period finishes, the infrastructure should be transferred back to the public sector. However, governments are prompt to promote reconcessions by rebidding PPPs. The implications of reconcessioning a PPP that has undergone a previous concession have not been thoroughly considered in the literature. This study develops a System Dynamics (SD) model to examine the implications of reconcession in a national user-pay PPP program over the last 30 years aiming to provide a tool for decision-makers to understand the interrelations between reconcessions and the main PPP outcomes. This model explores the implications of periodical reconcessions by unravelling five asymmetries derived from reconcessions and assesses their impacts on longer average concession periods, higher project expenditures, higher user fees, lower uncertainty and reduced demand volatility. Findings have important implications for PPP policymakers who should thoroughly consider reconcessions implications on PPP program expenses and reliance on user revenues that may trigger erosion of social legitimacy and potential program closure due to user fee increases, demand elasticity, and demand volatility when making decisions about reconcessions in PPP programs. The model’s reliability is tested based on a national toll road PPP program including more than 60 projects.

Advancing Sustainable Infrastructure Management: Insights from System Dynamics

Rapid infrastructure growth in developing countries has intensified environmental challenges due to cost-prioritizing practices over sustainability. This study evaluates 21 identified sustainable-driving tools to improve the management of infrastructure throughout its life cycle, by interacting with 20 out of 36 key infrastructure system management variables (ISMVs). Using a systems thinking approach, a Sustainable Systems Dynamic Model (SSDM) is developed, comprising a nucleus representing the interconnected stages of the life cycle: planning and design (S1), procurement (S2), construction (S3), operation and maintenance (S4), and renewal and disposal (S5). The model incorporates a total of 12 balance (B) and 25 reinforcement (R) loops, enabling the visualization of critical interdependencies that influence the sustainability of the system. In addition, its analysis shows the interdependencies between variables and stages, demonstrating, for example, how the implementation of tools such as LCA, BIM, and Circular Economy principles in S1, or IoT and SHM in S4, significantly improve sustainability. A gap between theory and practice in the adoption of sustainable practices is identified, which is aggravated by the lack of knowledge in specific developing countries’ context. Hence, this study contributes to its closure by offering a model that facilitates the understanding of key interactions in infrastructure systems.

A System Dynamics Approach to Concession Period Optimization in Public-Private Partnerships (PPPs)

The success of public-private partnerships (PPPs) often depends on the duration of a project’s concession period. Although experts have invested considerable efforts into optimizing these periods, few have examined the complex, interactive effects of project-specific risks and bankability criteria on concession durations. To overcome this gap, this study develops a System Dynamics (SD) model that allows decision-makers to estimate the optimal concession period terms based on causal interactions between a project’s main risks and bankability. The model is applied using data from recent PPPs in Italy. The results show variable interest rates, inflation, operational expenditures (OPEX), and debt repayment mechanisms all uniquely impact the concession period.

A novel DEA-Tobit-SD assessment framework and application of provincial-level carbon emission embracing regional heterogeneity

Formulating tailored emission reduction policies for each Chinese province is crucial due to regional differences in carbon emission evolution patterns. This paper proposes a novel and comprehensive research framework that integrates data envelopment analysis (DEA), Tobit regression, and system dynamics (SD) model to analyze the influence factors and evaluate provincial emission reduction policies while considering regional differences. The DEA method assesses each province's development resource allocation and carbon emission efficiency. Based on the DEA results, each provinces’ key emission influencing factors can be derived combining with Tobit regression and sensitivity analysis of SD. Policies are then selected based on these factors to gauge their effectiveness. SD method is used to simulate carbon emissions under different policy scenarios in the future. The analysis results present obvious differences in resource allocation and regional characteristics among provinces. Qinghai's emission reduction potential has been preliminarily explored as an example. Energy structure, industry structure, energy intensity, forest coverage, and R&D input intensity are its main influencing factors for carbon emission. The forest carbon sink plays a significant role. The emission reduction of the integrated scenario is not a linear sum of all other scenarios. To ensure the completion of the neutralization goal, further adjustments to the long-term policy and extra measures are needed.

A System Dynamics Simulating Model of Groundwater Level Changes and Its Impacts on Land Uplift

The global effort to transition from fossil fuels to renewable energy sources results in the closure of surface and underground coal mines, subsequently affecting the environment, society, and the economy. Land uplift is one of the challenges since the coal mine closure in Germany. Using computational modeling and simulation helps understand the long-term land uplift behavior related to variables’ complexity and dynamic interactions of groundwater level changes. This study proposes a system dynamics (SD) model to study the complexity of variables and consider the upstream groundwater impacts on land uplift in abandoned mines in the Ruhr district in Germany. The proposed model utilized causal loop diagrams (CLD) and stock flow diagrams (SFD) to provide a conceptual and simulation framework. The individual and combined effects of increasing rainfall, groundwater level, effective stress, porosity, porewater pressure, total stress, etc., on land uplift are studied. The model was evaluated by analyzing different scenarios based on stochastic and random variables in the Eastern Ruhr region of Germany. The scenario result showed that with a 10% increase in groundwater level, the land uplift increased by about + 6%, and a rise of 200% in total stress caused an 18% increase in effective stress, which decreased the land uplift by about 20%. Additionally, decreasing porosity leads to reduced land uplift. The land uplift trend was very high in the first 15 years after mine closure. Some scenarios are considered to delay and decrease the land uplift in the long-term.

KalmanFormer: using transformer to model the Kalman Gain in Kalman Filters.

Tracking the hidden states of dynamic systems is a fundamental task in signal processing. Recursive Kalman Filters (KF) are widely regarded as an efficient solution for linear and Gaussian systems, offering low computational complexity. However, real-world applications often involve non-linear dynamics, making it challenging for traditional Kalman Filters to achieve accurate state estimation. Additionally, the accurate modeling of system dynamics and noise in practical scenarios is often difficult. To address these limitations, we propose the KalmanFormer, a hybrid model-driven and data-driven state estimator. By leveraging data, the KalmanFormer promotes the performance of state estimation under non-linear conditions and partial information scenarios. The proposed KalmanFormer integrates classical Kalman Filter with a Transformer framework. Specifically, it utilizes the Transformer to learn the Kalman Gain directly from data without requiring prior knowledge of noise parameters. The learned Kalman Gain is then incorporated into the standard Kalman Filter workflow, enabling the system to better handle non-linearities and model mismatches. The hybrid approach combines the strengths of data-driven learning and model-driven methodologies to achieve robust state estimation. To evaluate the effectiveness of KalmanFormer, we conducted numerical experiments in both synthetic and real-world dataset. The results demonstrate that KalmanFormer outperforms the classical Extended Kalman Filter (EKF) in the same settings. It achieves superior accuracy in tracking hidden states, demonstrating resilience to non-linearities and imprecise system models.

Incorporating Deep Learning Into System Dynamics: Amortized Bayesian Inference for Scalable Likelihood‐Free Parameter Estimation

ABSTRACTEstimating parameters and their credible intervals for complex system dynamics models is challenging but critical to continuous model improvement and reliable communication with an increasing fraction of audiences. The purpose of this study is to integrate Amortized Bayesian Inference (ABI) methods with system dynamics. Utilizing Neural Posterior Estimation (NPE), we train neural networks using synthetic data (pairs of ground truth parameters and outcome time series) to estimate parameters of system dynamics models. We apply this method to two example models: a simple Random Walk model and a moderately complex SEIRb model. We show that the trained neural networks can output the posterior for parameters instantly given new unseen time series data. Our analysis highlights the potential of ABI to facilitate a principled, scalable, and likelihood‐free inference workflow that enhance the integration of models of complex systems with data. Accompanying code streamlines application to diverse system dynamics models.

System Dynamics Modeling of Caries Severity States in Long-Term Care

Dental caries among long-term care (LTC) residents is a persistent and complex problem driven by social and structural factors. Systems thinking may be useful in considering novel approaches to reducing disease. This study aimed to develop a system dynamics model to simulate the progression of dentate older adults in LTC through caries severity states and estimate the effects of 3 intervention scenarios on the progression of caries: preventive topical fluoride (TF), arrest of caries with silver diamine fluoride (SDF), and a combination of TF and SDF. Dentate older adults in LTC were categorized into 4 caries severity states by their number of untreated carious lesions. The model assumed that changes in severity states were consistent with incidence rates reported in the literature and available billing data for dental care and that individuals move in and out of the system by entering and exiting the facility or experiencing edentulism. For all scenarios, the proportion of dentate older adults in LTC with 1 or more untreated lesions stays stable, the distribution of disease shifts from a high severity state, and the system approaches equilibrium after 4 y. The TF intervention predicts minimal impacts on decreasing the proportion of dentate older adults with 1 or more untreated lesions (2.5% decrease), while the SDF intervention and the combination interventions were most disruptive. There was a 29.6% and 33.6% decrease, respectively. Given the specific population dynamics in LTC, these findings suggest that long-term (greater than 4 y) interventions should be designed to address both the management of existing lesions and their incidence. This system dynamics model allows researchers to render institution-specific data points from LTCs to estimate the effects of proposed interventions at the respective site.

Insights from a codesigned dynamic modelling study of child and adolescent obesity in Australia

IntroductionChild and adolescent obesity is associated with a range of immediate health issues and influences obesity in adulthood. The complex nature of health determinants that contribute to obesity makes it...

Modeling time-varying spatial accessibility to healthcare: A system dynamic approach.

Spatial accessibility to healthcare is essential for policymakers to identify health disparities and develop targeted interventions. Current modeling approaches poorly capture temporal dynamics of contributing factors, and few have represented dynamic interactions among these factors. Further, validating current models is hindered by data limitations and methodological challenges. To fill these gaps, I propose a new modeling approach that produces time-varying and empirically validated measures of healthcare accessibility. Specifically, I develop a system dynamic model to represent interplay between population demand and healthcare supply over time. The model simultaneously estimated people's potential accessibility to healthcare and their utilization over time and space, then validated these estimates with actual hospitalization data. To illustrate the model's practical use, a working prototype was constructed to estimate and validate healthcare accessibility by day and by ZIP code in Florida, USA, during a disease outbreak in 2022-23. The results indicate that system dynamic modeling offers a robust framework for monitoring healthcare accessibility fluctuations across regions and time periods, thereby guiding the development of timely and targeted interventions to reduce disparities.

Case study of smoke and heat control systems in high-rise buildings

New European technical regulations impose the need for Computational Fluid Dynamics (CFD) modeling of smoke and heat control systems for buildings with a height greater than 60 m. Consequently, this study presents CFD simulations in this context, focusing on two essential constructive elements in ensuring a safe evacuation of users from high-rise buildings, interconnected with each other: the smoke enclosure and the stairwell. It is worthwhile to mention that the developed numerical model considered requirements of technical regulations in force and data from specialized literature, as well as non-conformities of some constructive elements with fire protection roles. The results show the importance of all constructive elements with active and passive protection roles in case of fire. These elements contribute to reducing the effects produced by fires, especially in the case of non-conformities of the smoke and heat control systems, respectively of the smoke barriers (smokeproof enclosure access and exit fire doors).