Search

For the accurate prediction of a complex system, determining how to model well it is essential. A classical simulation modeling method that abstracts causality between inputs and outputs utilizing knowledge such as physical laws or operating rules is widely used. However, it may cause a problem in reliability of the model's validity if data acquisition of the actual system is difficult. Machine learning, on the other hand, is a method to represent a correlation between one set of data and another. The model can be built using the big data of the target system. It has a limitation in that it is impossible to predict accurately using the learned model if the parameters or the operating rules are changed after the model is learned. In this paper, we propose a collaborative modeling method using big data-based machine learning and simulation modeling. Specifically, a hypothetical model can be constructed through a cellular automata model (simulation modeling), and parameters and functions necessary for a hypothetical model can be simulated by learning and applying an artificial neural network model (machine learning). This paper shows that the proposed method can be applied to the traffic model to predict traffic congestion in an unsteady state.

Since an optimal computing budget allocation (OCBA) approach maximizes the efficiency of the simulation budget allocation to correctly find the optimal solutions, various OCBA-based procedures, such as OCBA, OCBAm+, and MOCBA+, have been widely applied to solve simulation-based optimization problems. Recently, it has been found that the stochastic noise in a simulation model increases due to the increasing complexity of modern industrial systems. However, the OCBA approach may be inefficient for these practical problems. That is, it is very likely to waste a lot of budget on other candidates that are not truly optimal due to the abnormal simulation results, which occurs frequently in noisy environments. In this paper, we intuitively analyze the causes of this efficiency deterioration of the OCBA approach, and then a simple heuristic adjustment is proposed to enhance the noise robustness of the OCBA approach based on our analysis results. The proposed adjustment allows the OCBA approach to further consider the precision of the simulation results, thereby significantly reducing the wasted budget and increasing the efficiently. In addition, it can be applied to the existing allocation rules without modification and does not require additional computational costs. Many experimental results for the eight OCBA-based procedures clearly demonstrate the effectiveness of this adjustment. In particular, the results of practical problems emphasize its necessity.

This paper discusses two modeling and simulation (M&S) issues for defense cyber-physical systems (CPSs): 1) model-driven development via model interoperation and 2) simulation analysis between interoperated models. To clearly describe the network capabilities between the CPSs within network-centric warfare (NCW), we first modeled communication factors separately from the overall CPSs. The CPS models and the network model were considered discrete event systems, which were realized with DEVSim++ and ns-3, respectively. For the simulation of the two types of models, we interoperated them with high-level architecture-based middleware to preserve the independence of each one. The goal for NCW analysis is to find a balance between the computational, physical, and communication factors of the defense CPSs to achieve their missions within a networked environment. In the case study of network-centric ground warfare, the CPS models generate mission effectiveness to be analyzed according to realistic network conditions; and at the same time, the network model enables the analysis of the network performance given validated combat scenarios. We expect this paper will help simulation experts interactively analyze computational and physical capabilities of the CPSs as well as communication effects between the CPSs.

Motivated by the increasing practical needs for simulation optimization of modern industrial systems, this paper proposes an efficient ranking and selection (R&S) procedure for selecting the best-simulated design from a finite set of alternatives in the presence of large stochastic noise. To obtain the correct selection under a limited simulation budget, the proposed procedure sequentially allocates the budget to minimize the evaluated uncertainty values of the selection through a two-step process based on the existing uncertainty evaluation (UE) procedure. This two-step process reduces the inefficiency of the underlying UE procedure while keeping its high robustness to noise, thereby achieving improved the efficiency for the proposed procedure in a noisy environment. This improved efficiency is demonstrated in comparative experiments with other R&S procedures on several benchmark problems. In particular, the experimental results of three practical optimization problems emphasize the necessity of the proposed procedure.

Communication system in the network-centric warfare (NCW) has been analyzed from the perspective of the system of systems (SoS), which consists of a combat system and a network system so that the two reflect each other’s effects. However, this paradoxically causes a prolonged execution time. To solve this problem, this paper proposes an advanced integrated modeling method for the communication analysis in the NCW via the transformation of the SoS, which reduces the simulation execution time while ensuring the accuracy of the communication effects. The proposed models mainly cover interentity traffic and intraentity mobility developed in the form of feed-forward neural networks to guarantee two-way interactions between the combat system and the network system. Because they are characterized as discrete events, the proposed models are designed with the discrete-event system specification (DEVS) formalism. The experimental results show that the proposed transformation reduced an error by 6.40% compared to the existing method and reduced the execution time 3.78-fold compared to the SoS-based NCW simulation.

Simulation of a system-of-systems (SoS) model, which consists of a combat model and a network model, has been used to analyze the performance of network-centric warfare in detail. However, finding the combat model parameters satisfying the required combat power using simulation can take a long time for two reasons: (1) the prolonged execution time per simulation run and (2) the enormous number of simulation runs. This paper proposes a simulation-based optimization method for the SoS-based simulation model to overcome these problems. The method consists of two processes: (1) the transformation of the SoS-based model into an integrated model using the neural network to reduce the execution time and (2) the optimization of the integrated model using the genetic algorithm with ranking and selection to decrease the number of simulation runs. The experimental result reveals that the proposed method significantly reduced the time for finding the optimal combat parameters with an acceptable level of accuracy.

Simulation analysis has been performed for simulation experiments of all possible input combinations as a “what-if” analysis, which causes the simulation to be extremely time-consuming. To resolve this problem, this paper proposes a multi-fidelity modeling framework for enhancing simulation speed while minimizing simulation accuracy loss. A target system for this framework is a discrete event dynamic system. The dynamic property of the system facilitates the development of variable fidelity models for the target system due to its high computational cost; and the discrete event property allows for determining when to change the fidelity within a simulation scenario. For formal representation, the paper defines several key concepts such as an interest region, a fidelity change condition, and a selection model. These concepts are integrated into the framework to allow for the achievement of a condition-based disjunction of high- and low-fidelity simulations within a scenario. The proposed framework is applied to two case studies: unmanned underwater and urban transportation vehicles. The results show that simulation speed increases at least 1.21 times with a 5% accuracy loss. We expect that the proposed framework will resolve a computationally expensive problem in the simulation analysis of discrete event dynamic systems.