Abstracts Track 2021

Area 1 - Simulation Tools and Platforms
Area 2 - Complex Systems Modeling and Simulation
Area 3 - Application Domains

Area 1 - Simulation Tools and Platforms

Nr: 10
Title:

Virtual Reality the Standard for Training across Industries
Authors:

Aida Otaola
Abstract: Workforce training continues to be a key area to help achieve some of the main targeted goals for the military and other high-risk industries: Implement the most significant savings opportunities in the most efficient manner possible. The demand for more realistic and innovative ways for training is increasing, where safety and knowledge retention are becoming even more important elements in today’s social and economic environment. For instance, the possibility to train several units based in different locations but in the same large-scale virtual environment. This improves the individual, team and unit-level performance in emergency and conflict situations, but also reduces the investment in physical replicas and even the expenses of bringing the units together to one location for training. Virtual Reality is establishing itself as the standard for training across industries. The abstract will show the potential of virtual reality to accelerate Innovation, revolution the education and improve workforce training with different real use cases. Use case 1: GE Hitachi is introducing large-scale, multi-user VR simulators in their training programmes. Through Virtualware Immersive Room (VIROO), they can simulate complex and highly skilled activities that require top-level expertise and training, as well as outstanding coordination between the different parties involved.The first simulator developed is aimed to train workers in fuel movement operations, allowing multiple users to collaborate in a refuel floor simulated nuclear environment. By means of a physical replica of the refuelling mast, an indispensable tool for carrying out fuel movement operations, users can carry out fuel movement operations as if they were doing so in reality. Use Case 2: The Spanish Ministry of Defense (MoD) is shaping the training processes modernization of the medical staff of the Military Health Corps in up to three highly complex environments related to (Chemical, biological, radiological, nuclear) CBRN protocols through a warehouse-scale multi-user VR simulator. Through a warehouse-scale multi-user VR simulator deployed in the VR innovation tool VIROO the medical staff at the Spanish Military Health School (EMISAN), can be prepared in interactive, realistic training sessions. It enables the Spanish MoD to reduce cost, risk and time in their learning and training processes. This is the first of the next generation of simulators that will be part of a wider training program for the Spanish Health Forces. Virtual reality technologies offer numerous exciting opportunities for an education system that can be instrumental in solving its key challenges and bringing to a new level of quality. Use case 3: The Technological University of El Retoño (UTR) is a non-profit public higher education institution located in El Retoño, Aguascalientes, Mexico. As a leading University in Mexico, UTR is committed to create the world’s future leaders with the latest and most innovative learning technologies. To make it possible they have joined forces with Virtualware to take traditional education to the next level, and make learning more relevant, resilient, and effective under VIROO™, a pedagogical and technological VR tool that promotes the design and delivery of didactic strategies that are more agile, effective, engaging, and enjoyable than traditional approaches, that foster learning and the development of skills as well as creative and exponential thinking.
Nr: 13
Title:

COVID-19 Simulated SIR Models via Announced Data: How Vaccinations Will Effect
Authors:

Alper Bilal
Abstract: Summary: The COVID-19 outbreak was announced as a pandemic by World Health Organization (WHO) on March 11, 2020. The first COVID-19 case was also confirmed at the same date in Turkey. The aim of this article is to simulate mathematical models of COVID-19 for Turkey (TR), The United Kingdom (GBR) and Germany (DEU), compare different scenarios in terms of interventions and to evaluate the impact on healthcare systems. Besides reflecting the possible effects of vaccinations is an additional interest for this article. Three (3) scenarios have been evaluated with the SIR Model: (1) Turkey with gradually increasing government interventions, followed by loosening on those government interventions (2) UK with loosened interventions at the beginning and afterwards gradually increasing and moreover constant government interventions, (3) Germany with gradually increasing and constant government interventions. Possible effects of vaccinations are also reflected on the models. The results show the critical importance of early and effective interventions. Loosening interventions will end up with high number of new cases. In terms of a wide spread of vaccination will be a good option for ending the pandemics. Although the number of casualties are more than Turkey and Germany, UK will be the first to end the Pandemics applying a more number of doses of vaccination.

Area 2 - Complex Systems Modeling and Simulation

Nr: 6
Title:

A Novel Method to Simplify the Simulation of Complex Stochastic Biochemical Reaction Networks
Authors:

Vincent Wagner and Nicole Radde
Abstract: The Chemical Master Equation (CME) is a standard approach toChemical Master Equation, Gillespie Algorithm, Model Reduction, Catalysts. model biochemical reaction networks. The CME consists of a system of linear differential equations, in which each state corresponds to a possible configuration of the reaction system, and the solution describes a time-dependent probability distribution over all configurations. The CME is a time-continuous Markov process over a discrete state space that can be visualized by a state transition graph. Nodes of this graph correspond to possible configurations, and each edge represents a reaction and is labeled with the respective reaction propensity. The Stochastic Simulation Algorithm (SSA) is a method to simulate sample paths from this stochastic process. Both approaches are only applicable for small systems, characterized by few reactions and small numbers of molecules. For larger systems, the CME is computationally intractable due to a large number of possible configurations, and the SSA suffers from high reaction propensities. In fact, it can be tedious to even capture the complete set of configurations and all possible reactions. In our study, we focus particularly on reactions that are driven by catalytic molecules, including for example post-transcriptional modifications of proteins such as phosphorylation reactions, epigenetic changes such as DNA methylation, and mRNA transcription. We show that these complex systems can analogously be described by a concatenation of simpler systems. This simplification is achieved by simulating one catalyst molecule at a time instead of all molecules together. The advantages of our approach are: (i) It is not necessary to define the state transition graph of the entire system. (ii) Intractable state transition graphs are replaced by a concatenation of much simpler graphs, resulting in lower-dimensional linear differential equations for the CME approach. (iii) The implementation of the SSA is considerably simplified. We show the validity of our approach by applying it to two test-bed reaction systems, degradation of a molecule modeled with simple first-order kinetics, and methyltransferase-mediated DNA methylation. For both systems, we show that simulating the catalyst molecules one after another is analytically equivalent to solving the full systems. We furthermore discuss conditions under which our approach is applicable to a larger bandwidth of systems.
Nr: 14
Title:

Data-driven Agent-based Model Building for Animal Movement through Exploratory Data Analysis
Authors:

David Butts, Noelle Thompson, Sonja Christensen, David Williams and Michael Murillo
Abstract: In data-starved fields, models that are chosen for convenience often require assumptions potentially making them incapable of describing realistic behaviors; however, higher-quality data can be used to develop new models that are able to reproduce more realistic patterns. Fields that study features of animal movement tend to be data-starved due to the large monetary and time costs required to tag and track animals for extended periods of time. The data available to researchers consists of the positions of multiple tagged animals recorded at a fixed time interval over the course of a total study period. Even with the data constraint, diverse models have been developed to study a variety of features of animal movement, including migration, land use, resource use, and other processes. Many of these models are based on random walks that use Gaussian noise, for example state-space models and hidden-Markov models, or heavy-tailed distributions resulting in Levy flights. When examining real movement data, these assumptions are not always valid on all spatial and time scales leading to a need for new models. We provide methods for developing and training more realistic agent-based models of animal movement by taking a data-driven approach. By performing an exploratory data analysis (EDA) on animal location data, our data will lead us to a model that will best reproduce the patterns we observe. In our EDA, we examine distributions of positions, calculate the autocorrelation of the movement data, use Fourier analysis, calculate mean-squared displacements, and test for correlations. Large changes in the temporal evolution are present in our movement data, creating a non-parametric trend. We introduce the fused-lasso regression-analysis method as a method for identifying these behaviors through a non-parametric fit that is sensitive to discontinuities. Movements additionally show correlations and are not Gaussian or heavy-tailed distributed. We introduce the copula and kernel-density estimates as methods for approximating the coordinate-system-independent movement correlations from the marginal location differences represented in our data set, allowing us to create non-Gaussian noise terms. With the insights gained from our EDA, we create a Langevin model that describes the movements of an individual animal that features non-Gaussian noise and incorporates multiple movement patterns. This model is extended to an agent-based model that describes the movements of groups of multiple deer. We compare our Langevin model to three models built on different assumptions that results in order-of-magnitude differences in the amount of area covered by an animal. Using our agent-based model, we simulate three groups of deer with parameters sampled from our data to illustrate the amount of area covered by deer in our model and the ability for groups to overlap in space. This overlapping behavior is an important feature in modeling processes driven by interactions between animals. While we apply our data-driven methods to animal movement, they are widely applicable to developing and training models from many data sources.

Area 3 - Application Domains

Nr: 15
Title:

Discrete Event Simulation Model for Human Resources Planning in Dental Caries Prevention among Primary School Children
Authors:

Maria Hajlasz and Bożena Mielczarek
Abstract: Dental caries is a disease that affects more than half of the world's population and can cause many complications. However, properly conducted prevention gives very good results. In order to provide widespread access to prophylactic services, it is important to properly plan human resources. Literature studies suggest there is a gap in the area of application of discrete event simulation (DES) method to support management decision making in tooth decay prevention. The purpose of this study is to determine the need for dental hygienist and dentist care necessary to provide preventive services in a primary school. In order to better understand the medical aspects of the research, the authors consulted experts and conducted a survey. A group of 24 respondents identified tooth decay among children as serious healthcare problem. In addition, it was noted that dental hygienist should be more involved in the prevention of tooth decay. The survey results were elaborated and then again consulted with doctors. Once validated, they were used as input to the simulation model. DES is a great tool to support decision making in planning the size and structure of human resources. We developed a DES model for a sample elementary school, from kindergarten (class “0”) to the eighth grade, with four classes in each school year. The model was fed with the sample data. A simulation of the full period of education in the sample school was carried out and statistics were collected. As part of the preventive measures of caries, the actions of a doctor and a dental hygienist were taken into account. Two experiments were conducted to test the conceptual assumptions of the model. The effect of the number and frequency of visits to the dentist and dental hygienist on oral health was studied. In experiment 1, which was the baseline experiment, we assumed that there were two visits to specialists per school year (18 in total). In experiment 2, we assumed that there was one visit to specialists per school year (nine in total). Each visit to specialists results in an update of the values of the adopted indicators. Each student starts school with a randomly assigned risk of caries and a condition of caries. The risk is given as a percentage, and the condition of caries as the number of teeth with caries. The preliminary results obtained show that by means of a DES, the number of specialists necessary for the prevention of caries in primary school can be determined. To provide each student with two visits per year to each specialist, a commitment of human resources of: 123 hours per month (doctor) and 6 hours per month (hygienist). The preliminary results were promising and revealed that the availability of preventive services for dental caries had an impact on oral health among primary school children. We found that simulation is a valuable tool in supporting resource planning decisions for dental prevention programs. We presented the concept of using the DES methodology for human resource planning for the prevention of tooth decay in primary school children.