생체의 모델링과 Simulation

MRA++: Scheduling and data placement on MapReduce for heterogeneous environments

SEMSim Cloud Service: Large-Scale Urban Systems Simulation in the Cloud

The computational cost of predicting wildland fire spread across large, diverse landscapes is significant using current models, which limits the ability to use simulations to develop mitigation strategies or perform forecasting. This paper presents a machine learning approach to estimate the time-resolved spatial evolution of a wildland fire front using a deep convolutional inverse graphics network (DCIGN). The DCIGN was trained and tested for wildland fire spread across simple homogeneous landscapes as well as heterogeneous landscapes having complex terrain. Data sets for training, validation, and testing were created using computational models. The model for homogeneous landscapes was based on a rate of spread from the model of Rothermel, while heterogeneous spread was modeled using FARSITE. Over 10,000 model predictions were made to determine burn maps in 6 h increments up to 24 h after ignition. Overall the predicted burn maps from the DCIGN-based approach agreed with simulation results, with mean precision, sensitivity, F-measure, and Chan–Vese similarity of 0.97, 0.92, 0.93, and 0.93, respectively. Noise in the input parameters was found to not significantly impact the DCIGN-based predictions. The computational cost of the method was found to be significantly better than the computational model for heterogeneous spatial conditions where a reduction in simulation time of $$10^{2}{-}10^{5}$$
 was observed. In addition, the DCIGN-based approach was shown to be capable of predicting burn maps further in the future by recursively using previous predictions as inputs to the DCIGN. The machine learning DCIGN approach was able to provide fire spread predictions at a computational cost three orders of magnitude less than current models.

Wildland Fire Spread Modeling Using Convolutional Neural Networks

Recently, most healthcare organizations focus their attention on reducing the cost of their supply chain management (SCM) by improving the decision making pertaining processes' efficiencies. The availability of products through healthcare SCM is often a matter of life or death to the patient; therefore, trial and error approaches are not an option in this environment. Simulation and modeling (SM) has been presented as an alternative approach for supply chain managers in healthcare organizations to test solutions and to support decision making processes associated with various SCM problems. This paper presents and analyzes past SM efforts to support decision making in healthcare SCM and identifies the key challenges associated with healthcare SCM modeling. We also present and discuss emerging technologies to meet these challenges.

/pdf/simulation-and-modeling-efforts-to-support-decision-making-5adnu5nqwj.pdf

Simulation and Modeling Efforts to Support Decision Making in Healthcare Supply Chain Management

Simulation software is a special category of software that is used by researchers or field experts. As the Service-Oriented Architecture gains its popularity, it is desirable to make simulation software, which originally is accessible only on a single machine, as a service that can be accessed by researchers and field experts over the Internet. Towards this goal, this paper provides a specification for simulation software as a service (SimSaaS) and service-oriented experiment in order to support automatic deployment of simulation services for carrying out experiments. We also present an architecture that provides an integrated experiment development environment which helps researchers to focus on the design of their experiments.

/pdf/simulation-software-as-a-service-and-service-oriented-1h0rz1p473.pdf

Simulation software as a service and Service-Oriented simulation experiment

Particle filters are important techniques to support data assimilation for large-scale spatial temporal simulation systems Distributed particle filters improve the performance of particle filtering by distributing particles to multiple processing units (PUs) While different resampling algorithms have been developed for distributed particle filters, less research has been conducted to investigate how to route particles among the PUs after resampling in effective and efficient manners This paper develops particle routing policies in distributed particle filters with both the centralized resampling and the distributed resampling The developed routing policies are evaluated from the aspects of the communication cost and the data assimilation accuracy based on an application of data assimilation for large-scale wildfire spread simulations

Particle Routing in Distributed Particle Filters for Large-Scale Spatial Temporal Systems

MapReduce is a domain-independent programming model for processing data in a highly parallel fashion. With MapReduce, parallel computing can be automatically carried out in large-scale commodity machines. This paper presents a method that utilizes the parallel and distributed processing capability of Hadoop MapReduce for particle filter-based data assimilation in wildfire spread simulation. We parallelize the sampling and weight computation steps of the particle filtering algorithm based on the MapReduce programming model. Experiment results show that our approach significantly increases the performance of particle filter-based data assimilation.

Cloud MapReduce for particle filter-based data assimilation for wildfire spread simulation

Simulation models rely on many parameters to model the structure and behavior of systems under study To achieve accurate simulation results, there is a need to develop methods to dynamically estimate the correct set of model parameters for a given simulation scenario In this paper, we present a method to dynamically estimate model parameters by assimilating real time data using Sequential Monte Carlo (SMC) methods We formulate the problem of single and multiple parameter estimations based on the context of wildfire spread simulation Preliminary results show that the developed method can be applied to parameter estimation in wildfire spread simulation to produce more accurate simulation results The complexity and difficulties in multiple parameter estimation are discussed as well

Fan Bai

Papers

Simulation software as a service and Service-Oriented simulation experiment

Particle Routing in Distributed Particle Filters for Large-Scale Spatial Temporal Systems

Cloud MapReduce for particle filter-based data assimilation for wildfire spread simulation

Towards parameter estimation in wildfire spread simulation based on sequential Monte Carlo methods