Agent-based simulation and child protection systems: Rationale, implementation, and verification

BACKGROUND AND OBJECTIVE

Simulation models are an important tool used in health care and other disciplines to support operational research and decision-making. In the child protection literature, simulation models are an under-utilized source of research evidence.

PARTICIPANTS AND SETTING

In this paper, we describe the rationale for and the development of an agent-based simulation of a child protection system in the US. Using the investigation, prevention service, and placement histories of 600,000 children served in an urban child welfare system, we walk the reader through the development of a prototype known as OSPEDALE.

METHODS

The governing equations built into OSPEDALE probabilistically simulate the onset of investigations. Then, drawing from empirical survival distributions, the governing equations trace the probability of subsequent interactions with the system (recurrence of maltreatment, service referrals, and placement) conditional on the characteristics of children, their assessed risk level, and prior child protection system involvement.

RESULTS

As an initial test of OSPEDALE's utility, we compare empirical admission counts with counts generated from OSPEDALE. Though the verification step is admittedly simple, the comparison shows that OSPEDALE replicates the empirical count of new admissions closely enough to justify further investment in OSPEDALE.

CONCLUSIONS

Management of public child protection systems is increasingly research evidence-dependent. The emphasis on research evidence as a decision-support tool has elevated evidence acquired through randomized clinical trials. Though important, the evidence from clinical trials represents only one type of research evidence. Properly specified, simulation models are another source of evidence with real-world relevance.

Investigating the first stage of the COVID-19 pandemic in Ukraine using epidemiological and genomic data

The novel coronavirus SARS-CoV-2 was first detected in China in December 2019 and has rapidly spread around the globe. The World Health Organization declared COVID-19 a pandemic in March 2020 just three months after the introduction of the virus. Individual nations have implemented and enforced a variety of social distancing interventions to slow the virus spread, that had different degrees of success. Understanding the role of non-pharmaceutical interventions (NPIs) on COVID-19 transmission in different settings is highly important. While most such studies have focused on China, neighboring Asian counties, Western Europe, and North America, there is a scarcity of studies for Eastern Europe. The aim of this epidemiological study is to fill this gap by analyzing the characteristics of the first months of the epidemic in Ukraine using agent-based modelling and phylodynamics. Specifically, first we studied the dynamics of COVID-19 incidence and mortality and explored the impact of epidemic NPIs. Our stochastic model suggests, that even a small delay of weeks could have increased the number of cases by up to 50%, with the potential to overwhelm hospital systems. Second, the genomic data analysis suggests that there have been multiple introductions of SARS-CoV-2 into Ukraine during the early stages of the epidemic. Our findings support the conclusion that the implemented travel restrictions may have had limited impact on the epidemic spread. Third, the basic reproduction number for the epidemic that has been estimated independently from case counts data and from genomic data suggest sustained intra-country transmissions.

Neighbourhoods and oral health: Agent-based modelling of tooth decay

This research used proof of concept agent-based models to test various theoretical mechanisms by which neighbourhoods may influence tooth decay in adults. Theoretical pathways were constructed using existing literature and tested in two study areas in Sheffield, UK. The models found a pathway between shops and sugar consumption had the most influence on adult tooth decay scores, revealing that similar mechanisms influence this outcome in different populations. This highlighted the importance of the interactions between neighbourhood features and individual level variables in influencing outcomes in tooth decay. Further work is required to improve the accuracy and reliability of the models.

Harnessing adaptive novelty for automated generation of cancer treatments

Nanoparticles have the potential to modulate both the pharmacokinetic and pharmacodynamic profiles of drugs, thereby enhancing their therapeutic effect. The versatility of nanoparticles allows for a wide range of customization possibilities. However, it also leads to a rich design space which is difficult to investigate and optimize. An additional problem emerges when they are applied to cancer treatment. A heterogeneous and highly adaptable tumour can quickly become resistant to primary therapy, making it inefficient. To automate the design of potential therapies for such complex cases, we propose a computational model for fast, novelty-based machine learning exploration of the nanoparticle design space. In this paper, we present an evolvable, open-ended agent-based model, where the exploration of an initially small portion of the given state space can be expanded by an ongoing generation of adaptive novelties, whenever the simulated tumour makes an adaptive leap. We demonstrate that the nano-agents can continuously reshape themselves and create a heterogeneous population of specialized groups of individuals optimized for tracking and killing different phenotypes of cancer cells. In the conclusion, we outline further development steps so this model could be used in real-world research and clinical practice.

Agent-based scenarios comparison for assessing fuel-switching investment in long-term energy transitions of the India's industry sector

This paper presents the formulation and application of a novel agent-based integrated assessment approach to model the attributes, objectives and decision-making process of investors in a long-term energy transition in India's iron and steel sector. It takes empirical data from an on-site survey of 108 operating plants in Maharashtra to formulate objectives and decision-making metrics for the agent-based model and simulates possible future portfolio mixes. The studied decision drivers were capital costs, operating costs (including fuel consumption), a combination of capital and operating costs, and net present value. Where investors used a weighted combination of capital cost and operating costs, a natural gas uptake of ~12PJ was obtained and the highest cumulative emissions reduction was obtained, 2 Mt CO₂ in the period from 2020 to 2050. Conversely if net present value alone is used, cumulative emissions reduction in the same period was lower, 1.6 Mt CO₂, and the cumulative uptake of natural gas was equal to 15PJ. Results show how the differing upfront investment cost of the technology options could cause prevalence of high-carbon fuels, particularly heavy fuel oil, in the final mix. Results also represent the unique heterogeneity of fuel-switching industrial investors with distinct investment goals and limited foresight on costs. The perception of high capital expenditures for decarbonisation represents a significant barrier to the energy transition in industry and should be addressed via effective policy making (e.g. carbon policy/price).

Long-term application potential of urban air mobility complementing public transport: an upper Bavaria example

In this paper, the required models and methods to analyze and quantify the potential demand for urban air mobility (UAM) complementing public transport and possible impacts were defined and applied to the Munich Metropolitan region. An existing agent-based transport model of the study area were used and extended to cover socio-demographic changes up to the year 2030 and intermodal UAM services. An incremental logit model for UAM was derived to simulate demand for this new mode. An airport access model was developed as well. Three different UAM networks with different numbers of vertiports were defined. Sensitivity studies of ticket fare and structure, flying vehicle cruise speed, passenger process times at vertiports and different Urban Air Mobility networks sizes were performed. For the reference case, UAM accounts for a modal share of 0.5%. The absolute UAM demand is concentrated on very short routes; hence, UAM vehicle flight speed variation shows low UAM demand impacts. Kilometer-based fare, number of UAM vehicles per vertiport and passenger process times at vertiports show a significant impact on UAM demand.

An agent-based computational framework for simulation of global pandemic and social response on planet X

The increase in readily available computational power raises the possibility that direct agent-based modeling can play a key role in the analysis of epidemiological population dynamics. Specifically, the objective of this work is to develop a robust agent-based computational framework to investigate the emergent structure of Susceptible-Infected-Removed/Recovered (SIR)-type populations and variants thereof, on a global planetary scale. To accomplish this objective, we develop a planet-wide model based on interaction between discrete entities (agents), where each agent on the surface of the planet is initially uninfected. Infections are then seeded on the planet in localized regions. Contracting an infection depends on the characteristics of each agent-i.e. their susceptibility and contact with the seeded, infected agents. Agent mobility on the planet is dictated by social policies, for example such as "shelter in place", "complete lockdown", etc. The global population is then allowed to evolve according to infected states of agents, over many time periods, leading to an SIR population. The work illustrates the construction of the computational framework and the relatively straightforward application with direct, non-phenomenological, input data. Numerical examples are provided to illustrate the model construction and the results of such an approach.

Mathematical modelling for health systems research: a systematic review of system dynamics and agent-based models

BACKGROUND

Mathematical modelling has been a vital research tool for exploring complex systems, most recently to aid understanding of health system functioning and optimisation. System dynamics models (SDM) and agent-based models (ABM) are two popular complementary methods, used to simulate macro- and micro-level health system behaviour. This systematic review aims to collate, compare and summarise the application of both methods in this field and to identify common healthcare settings and problems that have been modelled using SDM and ABM.

METHODS

We searched MEDLINE, EMBASE, Cochrane Library, MathSciNet, ACM Digital Library, HMIC, Econlit and Global Health databases to identify literature for this review. We described papers meeting the inclusion criteria using descriptive statistics and narrative synthesis, and made comparisons between the identified SDM and ABM literature.

RESULTS

We identified 28 papers using SDM methods and 11 papers using ABM methods, one of which used hybrid SDM-ABM to simulate health system behaviour. The majority of SDM, ABM and hybrid modelling papers simulated health systems based in high income countries. Emergency and acute care, and elderly care and long-term care services were the most frequently simulated health system settings, modelling the impact of health policies and interventions such as those targeting stretched and under resourced healthcare services, patient length of stay in healthcare facilities and undesirable patient outcomes.

CONCLUSIONS

Future work should now turn to modelling health systems in low- and middle-income countries to aid our understanding of health system functioning in these settings and allow stakeholders and researchers to assess the impact of policies or interventions before implementation. Hybrid modelling of health systems is still relatively novel but with increasing software developments and a growing demand to account for both complex system feedback and heterogeneous behaviour exhibited by those who access or deliver healthcare, we expect a boost in their use to model health systems.

Simulation tools for assessment of tick suppression treatments of Rhipicephalus (Boophilus) microplus on non-lactating dairy cattle in Puerto Rico

BACKGROUND

The southern cattle fever tick (SCFT), Rhipicephalus (Boophilus) microplus, remains endemic in Puerto Rico. Systematic treatment programmes greatly reduced and even eradicated temporarily this tick from the island. However, a systemic treatment programme that includes integrated management practices for livestock against SCFT remains to be established in the island. We describe a spatially-explicit, individual-based model that simulates climate-livestock-SCFT-landscape interactions. This model was developed as an investigative tool to aid in a research project on integrated management of the SCFT that took place in Puerto Rico between 2014 and 2017. We used the model to assess the efficacy of tick suppression and probability of tick elimination when applying safer acaricides at 3-week intervals to different proportions of a herd of non-lactating dairy cattle.

RESULTS

Probabilities of eliminating host-seeking larvae from the simulated system decreased from ≈ 1 to ≈ 0 as the percentage of cattle treated decreased from 65 to 45, with elimination probabilities ≈ 1 at higher treatment percentages and ≈ 0 at lower treatment percentages. For treatment percentages between 65% and 45%, a more rapid decline in elimination probabilities was predicted by the version of the model that produced higher densities of host-seeking larvae. Number of weeks after the first acaricide application to elimination of host-seeking larvae was variable among replicate simulations within treatment percentages, with within-treatment variation increasing markedly at treatment percentages ≤ 65. Number of weeks after first application to elimination generally varied between 30 and 40 weeks for those treatment percentages with elimination probabilities ≈ 1.

CONCLUSIONS

Explicit simulation of the spatial and temporal dynamics of off-host (host-seeking) larvae in response to control methods should be an essential element of research that involves the evaluation of integrated SCFT management programmes. This approach could provide the basis to evaluate novel control technologies and to develop protocols for their cost-effective use with other treatment methods.

Traffic accident risk assessment with dynamic microsimulation model using range-range rate graphs

Analysis of accidents that involve vehicles and pedestrians requires accurate reproduction of the dynamics of the vehicles and pedestrians immediately prior to and during the accident. In many cases, only centimeters and milliseconds separate survival from disaster, particularly when high-speed aggressive drivers and careless pedestrians are involved. In this paper we present a methodology for analyzing the dynamic interaction between drivers in conflict scenarios with pedestrians. We assess the safety of a traffic location's environment with a high-resolution, spatially explicit, dynamic agent-based simulation model - SAFEPED. Based on the resulting data, Range-Range Rate (R-RR) graphs are generated. These graphs provide compact, simple, and objective presentation of the dynamic interaction between vehicles and pedestrians. Significant traffic risk indicators such as Time-To-Collision, acceleration/deceleration rates, and minimal distances between vehicles and pedestrians are easily extracted from the R-RR graphs. These indicators can provide insights on particular traffic scenarios and can assist road planners and developers of traffic safety measures in understanding the dynamic behavior of drivers and pedestrians before and during a conflict scenario.

Size Matters: Metastatic Cluster Size and Stromal Recruitment in the Establishment of Successful Prostate Cancer to Bone Metastases

Prostate cancer (PCa) impacts over 180,000 men every year in the USA alone, with 26,000 patients expected to succumb to the disease ( cancer.gov ). The primary cause of death is metastasis, with secondary lesions most commonly occurring in the skeleton. Prostate cancer to bone metastasis is an important, yet poorly understood, process that is difficult to explore with experimental techniques alone. To this end we have utilized a hybrid (discrete-continuum) cellular automaton model of normal bone matrix homeostasis that allowed us to investigate how metastatic PCa can disrupt the bone microenvironment. Our previously published results showed that PCa cells can recruit mesenchymal stem cells (MSCs) that give rise to bone-building osteoblasts. MSCs are also thought to be complicit in the establishment of successful bone metastases (Lu, in Mol Cancer Res 4(4):221-233, 2006). Here we have explored the aspects of early metastatic colonization and shown that the size of PCa clusters needs to be within a specific range to become successfully established: sufficiently large to maximize success, but not too large to risk failure through competition among cancer and stromal cells for scarce resources. Furthermore, we show that MSC recruitment can promote the establishment of a metastasis and compensate for relatively low numbers of PCa cells seeding the bone microenvironment. Combined, our results highlight the utility of biologically driven computational models that capture the complex and dynamic dialogue between cells during the initiation of active metastases.

Lessons from a decade of individual-based models for infectious disease transmission: a systematic review (2006-2015)

Background

Individual-based models (IBMs) are useful to simulate events subject to stochasticity and/or heterogeneity, and have become well established to model the potential (re)emergence of pathogens (e.g., pandemic influenza, bioterrorism). Individual heterogeneity at the host and pathogen level is increasingly documented to influence transmission of endemic diseases and it is well understood that the final stages of elimination strategies for vaccine-preventable childhood diseases (e.g., polio, measles) are subject to stochasticity. Even so it appears IBMs for both these phenomena are not well established. We review a decade of IBM publications aiming to obtain insights in their advantages, pitfalls and rationale for use and to make recommendations facilitating knowledge transfer within and across disciplines.

Methods

We systematically identified publications in Web of Science and PubMed from 2006-2015 based on title/abstract/keywords screening (and full-text if necessary) to retrieve topics, modeling purposes and general specifications. We extracted detailed modeling features from papers on established vaccine-preventable childhood diseases based on full-text screening.

Results

We identified 698 papers, which applied an IBM for infectious disease transmission, and listed these in a reference database, describing their general characteristics. The diversity of disease-topics and overall publication frequency have increased over time (38 to 115 annual publications from 2006 to 2015). The inclusion of intervention strategies (8 to 52) and economic consequences (1 to 20) are increasing, to the detriment of purely theoretical explorations. Unfortunately, terminology used to describe IBMs is inconsistent and ambiguous. We retrieved 24 studies on a vaccine-preventable childhood disease (covering 7 different diseases), with publication frequency increasing from the first such study published in 2008. IBMs have been useful to explore heterogeneous between- and within-host interactions, but combined applications are still sparse. The amount of missing information on model characteristics and study design is remarkable.

Conclusions

IBMs are suited to combine heterogeneous within- and between-host interactions, which offers many opportunities, especially to analyze targeted interventions for endemic infections. We advocate the exchange of (open-source) platforms and stress the need for consistent “branding”. Using (existing) conventions and reporting protocols would stimulate cross-fertilization between research groups and fields, and ultimately policy making in decades to come.

Electronic supplementary material

The online version of this article (doi:10.1186/s12879-017-2699-8) contains supplementary material, which is available to authorized users.