Throughput,flow times,and service level in an unreliable assembly system

This paper considers an unreliable assembly network where different types of components are processed by two separate work centers before being merged at an assembly station. The operation complexity of the system is a result of finite inter-station buffers, uncertain service times, and random breakdowns that lead to blocking at the work centers and starvation at the assembly station. The objective of this study is to gain an understanding of the behavior of such systems so that we can find a way to maximize the system throughput while maintaining the required customer service level. By constructing appropriate Markov processes, we obtain the probability distribution of the production flow time and derive formulas for throughput, the loss probability of type-2 workpieces, and the mean flow time. We present expressions for average work-in-process (WIP) and study their monotone properties. Using the distribution of the flow time, a customer service level can be defined and computed. We then formulate a system optimization model that can be used to maximize the throughput while maintaining an acceptable service level.     

Dependence structure between LIBOR rates by copula method

This paper discusses the correlation structure between London Interbank Offered Rates (LIBOR) by using the copula function. We start from one simplified model of A. Brace, D. Gatarek, and M. Musiela (1997) and find out that the copula function between two LIBOR rates can be expressed as a sum of an infinite series, where the main term is a distribution function with Gaussian copula. Partial differential equation method is used for deriving the copula expansion. Numerical results show that the copula of the LIBOR rates and Gaussian copula are very close in the central region and differ in the tail, and the Gaussian copula approximation to the copula function between the LIBOR rates provides satisfying results in the normal situation.     

统计模拟在几何概率问题中应用的注解

设D为R²上的一个紧集, X 为D上的一个随机覆盖过程的统计量. 由于问题复杂, X的均 值、方差、分布函数均没有解析表达式. 统计模拟可以帮助我们找到它们的近似解. 为了在D上做统 计模拟, 需要D的代表点. 产生代表点的不同方法, 会影响统计模拟的结果. 若D不是一个矩形, 如 何选择合适的代表点至关重要. 文献中研究了一个在单位圆上的随机覆盖问题, 提出在单位圆上产生 代表点的四种方法, 并对这四种方法给予评估. 本文考虑两个随机圆的随机覆盖问题, 给出覆盖面积 的理论公式, 使比较四种产生代表点的方法有一个基准. 我们的研究结果和文献中的结论一致, 并发现 其中两种方法使覆盖面积均值的估计有偏, 且有较大的方差, 这是一个新的结果. 本文进一步指出覆 盖面积的分布可由 β 分布来拟合.     

Numerical Solution of Newtonian and Non-Newtonian Flows

This paper deals with the numerical solution of Newtonian and non-Newtonian flows. The flows are supposed to be laminar, viscous, incompressible and steady. The model used for non-Newtonian fluids is some variant of power-law. Governing equations in this model are incompressible Navier-Stokes equations. For numerical solution one could use artificial compressibility method with three stage Runge-Kutta finite volume method in cell centered formulation for discretization of space derivatives. Following cases of flows are solwed: flow through a bypass connected to main channel in 2D and 3D and non-Newtonian flow through branching channels in 2D. These results are presented for 2D and 3D case. This problem could have an application in the area of biomedicine. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Development of POD reduced-order model and its closure scheme for 2D Rayleigh–Bénard convection

Reduced-order model (ROM) based on proper orthogonal decomposition (POD) is a fast computational fluid dynamics (CFD) method and has been widely applied to pure flow or heat conduction problems in the past. In this paper, the typical 2D Rayleigh–Bénard convection (RBC) in a square cavity was set as a research target. Firstly, the POD-ROM of 2D RBC problem was constructed at Ra = 10⁷, Pr = 0.71. Combining with direct numerical simulation (DNS) databases, a closure model (CM) was then proposed to correct the evolution process of POD-ROM. Based on the proposed CM, we realized the prediction of flow evolution for a new flow case under the parameters different from that used to get its POD eigenmodes. It showed that the proposed POD-ROM with CM could be able to predict the dynamics of new flow cases. Moreover, the corresponding method proposed in the present study can be also easily extended to other types of flow-heat coupling problems, such as natural heat convection, etc.     

Coupling 3D and 1D fluid-structure interaction models for blood flow simulations

Three-dimensional (3D) simulations of blood flow in medium to large vessels are now a common practice. These models consist of the 3D Navier-Stokes equations for incompressible Newtonian fluids coupled with a model for the vessel wall structure. However, it is still computationally unaffordable to simulate very large sections, let alone the whole, of the human circulatory system with fully 3D fluid-structure interaction models. Thus truncated 3D regions have to be considered. Reduced models, one-dimensional (1D) or zero-dimensional (0D), can be used to approximate the remaining parts of the cardiovascular system at a low computational cost. These models have a lower level of accuracy, since they describe the evolution of averaged quantities, nevertheless they provide useful information which can be fed to the more complex model. More precisely, the 1D models describe the wave propagation nature of blood flow and coupled with the 3D models can act also as absorbing boundary conditions. We consider in this work the coupling of a 3D fluid-structure interaction model with a 1D hyperbolic model. We study the stability of the coupling and present some numerical results. (© 2006 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)     

Chaos in a dynamic model of urban transportation network flow based on user equilibrium states

In this study, we investigate the dynamical behavior of network traffic flow. We first build a two-stage mathematical model to analyze the complex behavior of network flow, a dynamical model, which is based on the dynamical gravity model proposed by Dendrinos and Sonis [Dendrinos DS, Sonis M. Chaos and social-spatial dynamic. Berlin: Springer-Verlag; 1990] is used to estimate the number of trips. Considering the fact that the Origin–Destination (O–D) trip cost in the traffic network is hard to express as a functional form, in the second stage, the user equilibrium network assignment model was used to estimate the trip cost, which is the minimum cost of used path when user equilibrium (UE) conditions are satisfied. It is important to use UE to estimate the O–D cost, since a connection is built among link flow, path flow, and O–D flow. The dynamical model describes the variations of O–D flows over discrete time periods, such as each day and each week. It is shown that even in a system with dimensions equal to two, chaos phenomenon still exists. A “Chaos Propagation” phenomenon is found in the given model.     

Periodic solutions of certain hybrid delay-differential equations and their corresponding difference equations

In this paper we use a method due to Carvalho (A method to investigate bifurcation of periodic solution in retarded differential equations, J. Differ. Equ. Appl. 4 (1998), pp. 17–27) to obtain conditions for the existence of nonconstant periodic solutions of certain systems of hybrid delay-differential equations. We first deal with a scalar equation of Lotka–Valterra type; then a system of two equations in two unknowns that could model the interactions of two identical neurons. It will be seen that such solutions are determined by solutions of corresponding difference equations. Another paper in which this method is used is by Cooke and Ladeira (Applying Carvalho's method to find periodic solutions of difference equations, J. Differ. Equ. Appl. 2 (1996), pp. 105–115).

We first state Carvalho's result.     

Bifurcation analysis of some forced Lu systems

Bifurcation behaviour of a forced Lu system is analyzed as the system parameter c and a forcing parameter F are varied. The Lu system belongs to a family of generalized Lorenz system. Members of this family are known to exhibit different types of chaotic attractors. Some of these attractors have been named Lorenz type L, Lu or Transition type T, Chen type T and Transverse 8 Type S. These different types of chaotic attractors are visually distinct when the parameters are widely separated. However, there is a need for identifying the precise point where transition from one type of chaotic attractor to another takes place. We identified signatures in the return map, which could be used for determining the point of transition and classifying the different types of chaotic attractors. These signatures helped to identify the point in coordinate space associated with such transitions. We find that such transitions take place when a chaotic attractor comes very close to a one-dimensional manifold on which the time derivatives of two of the variables is zero. We also find that just before coming to this point in coordinate space associated with the transition, the trajectory had approached, very closely, the equilibrium point at the origin.     

A generalized bi-criteria fuzzy integer flow sharing problem

The flow sharing problem is a class of techniques that can be used to find the optimal flow in a capacitated network, which realizes an equitable distribution of flows. This paper extends the integer flow sharing problem by considering fuzzy capacities and fuzzy weights such that the flux received at each sink node and the flow value through each arc are restricted to be multiples of some block unit. Fuzzy capacity describes the flexibility of the upper limit of flow value through each arc. Fuzzy weight represents the degree of satisfaction of the flux to a sink node. Our model has the two following criteria: to maximize the minimal degree of satisfaction among all of the fuzzy capacity constraints and to maximize the minimal degree of satisfaction among the fluxes to all of the sink nodes. Because an optimal flow pattern that simultaneously maximizes the two objectives is usually not feasible, we define non-domination in this setting and propose a pseudo-polynomial algorithm that finds some non-dominated flow patterns. Finally, a numerical example is presented to demonstrate how our algorithm works.     

Inference for the Number of Topics in the Latent Dirichlet Allocation Model via Bayesian Mixture Modeling

In latent Dirichlet allocation, the number of topics, T, is a hyperparameter of the model that must be specified before one can fit the model. The need to specify T in advance is restrictive. One way of dealing with this problem is to put a prior on T, but unfortunately the distribution on the latent variables of the model is then a mixture of distributions on spaces of different dimensions, and estimating this mixture distribution by Markov chain Monte Carlo is very difficult. We present a variant of the Metropolis–Hastings algorithm that can be used to estimate this mixture distribution, and in particular the posterior distribution of the number of topics. We evaluate our methodology on synthetic data and compare it with procedures that are currently used in the machine learning literature. We also give an illustration on two collections of articles from Wikipedia. Supplemental materials for this article are available online.     

A 3D non-linear k–ε turbulent model for prediction of flow and mass transport in channel with vegetation

The results from a 3D non-linear k–ε turbulence model with vegetation are presented to investigate the flow structure, the velocity distribution and mass transport process in a straight compound open channel and a curved open channel. The 3D numerical model for calculating flow is set up in non-orthogonal curvilinear coordinates in order to calculate the complex boundary channel. The finite volume method is used to disperse the governing equations and the SIMPLEC algorithm is applied to acquire the coupling of velocity and pressure. The non-linear k–ε turbulent model has good useful value because of taking into account the anisotropy and not increasing the computational time. The water level of this model is determined from 2D Poisson equation derived from 2D depth-averaged momentum equations. For concentration simulation, an expression for dispersion through vegetation is derived in the present work for the mixing due to flow over vegetation. The simulated results are in good agreement with available experimental data, which indicates that the developed 3D model can predict the flow structure and mass transport in the open channel with vegetation.     

The Bernoulli Integral for a Certain Class of Non‐Stationary Viscous Vortical Flows of Incompressible Fluid

It has been shown in our paper [1] that there is a wide class of 3D motions of incompressible viscous fluid which can be described by one scalar function dabbed the quasi‐potential. This class of fluid flows is characterized by three‐component velocity field having two‐component vorticity field; both these fields can depend of all three spatial variables and time, in general. Governing equations for the quasi‐potential have been derived and simple illustrative example of 3D flow has been presented. Here, we derive the Bernoulli integral for that class of flows and compare it against the known Bernoulli integrals for the potential flows or 2D stationary vortical flows of inviscid fluid. We show that the Bernoulli integral for this class of fluid motion possesses unusual features: it is valid for the vortical nonstationary motions of a viscous incompressible fluid. We present a new very nontrivial analytical example of 3D flow with two‐component vorticity which hardly can be obtained by any of known methods. In the last section, we suggest a generalization of the developed concept which allows one to describe a certain class of 3D flows with the 3D vorticity.     

FLOODING ON THE EEL RIVER: SYSTEM THEORETIC TIME SERIES,TRANSFER FUNCTION TIME SERIES,AND STRUCTURAL MODEL FORECASTS

Three models are developed to forecast precipitation, stream flow, and suspended sediment load for the Middle Fork Eel River basin near Dos Rios, California. The models are a structural model, a transfer function time series model, and a multivariate time series model based on the ideas of linear systems theory. Monthly observations from 1961 through 1970 were used to specify and estimate the models, and their performance was evaluated using 36 observations from 1971 through 1973. The comparisons show that all of the models are able to simulate the general trend of the data fairly well but that the two time series models capture the detail better. The system theoretic time series model fares the best, providing consistently better forecasts of all three series. We conclude that this model would be valuable for use in planning water supply systems in the Eel River basin.     

A two scale cellular automaton for flow dynamics modeling (2CAFDYM)

In this paper we propose a Two scale Cellular Automaton for Flow DYnamics Modeling (2CAFDYM) in a lowland region. Cells are terrain meshes with a predefined size, arranged in a bi-dimensional hexagonal lattice. The state of the cell consists of two scales: groundwater and surface water, in order to combine flows over saturated soil (Dunne flow) and flows exceeding the infiltration capacity (Hortonian flow). This allows for survey flood events and water resources. Each cell has intrinsic terrain attributes: altitude, soil type and land use. The obtained slopes are considered towards all the neighboring cells such that water flows simultaneously in multiple directions during the same time step. This helps us characterize laminar and turbulent flows. The model is subjected to climatic constraints: rainfall and temperature. The flow dynamics are regulated by mass conservation laws on hydraulic balance sheets (received, evaporated, infiltrated and drained water). Using Java Object Oriented Programming we have designed decision-aided software for the real-time monitoring of flow processes in 2D or 3D scenes through 2CAFDYM. We give some simulations for a basin in northern Morocco covering 34.3 km², including some areas that are potentially vulnerable to flooding. Digital terrain models, geological maps and satellite images are used to extract input data.     

A new integrated surface and subsurface flows model and its verification

A new coupled model for simulating surface and subsurface flows in a fully integrated way is presented. This model contains two sub-models; one is the 2D kinematic wave approximation of the Saint Venant’s equations used to model runoff, and another is Richards’ equation for variably saturated subsurface flow. In this model, boundary conditions (the conditions describing groundwater discharge at the land surface or surface water infiltration into the subsurface) could be eliminated through mathematic transformations of the governing equation of surface and subsurface flows. The solution of surface and subsurface flows could be simultaneous. And the surface domain and subsurface domain could be considered as a fully integrated domain. This approach naturally provides pressure and fluxes continuity along land surface. In order to assess this modelling approach, several classical validations, verification and application test cases are presented. For overland flow solely, the model is compared to an analytical solution and to commonly use hydrological models. The integrated model is then validated with a sandbox laboratory experiment and a soil column test. Finally, the effects of rainfall intensity, hydraulic conductivity of soils and initial bulk water content of soils to runoff and infiltration of a homogeneous soil slope are studied under different conditions.     

Sample selection bias in credit scoring models

One of the aims of credit scoring models is to predict the probability of repayment of any applicant and yet such models are usually parameterised using a sample of accepted applicants only. This may lead to biased estimates of the parameters. In this paper we examine two issues. First, we compare the classification accuracy of a model based only on accepted applicants, relative to one based on a sample of all applicants. We find only a minimal difference, given the cutoff scores for the old model used by the data supplier. Using a simulated model we examine the predictive performance of models estimated from bands of applicants, ranked by predicted creditworthiness. We find that the lower the risk band of the training sample, the less accurate the predictions for all applicants. We also find that the lower the risk band of the training sample, the greater the overestimate of the true performance of the model, when tested on a sample of applicants within the same risk band — as a financial institution would do. The overestimation may be very large. Second, we examine the predictive accuracy of a bivariate probit model with selection (BVP). This parameterises the accept–reject model allowing for (unknown) omitted variables to be correlated with those of the original good–bad model. The BVP model may improve accuracy if the loan officer has overridden a scoring rule. We find that a small improvement when using the BVP model is sometimes possible.     

Stability of the planar equilibrium solutions of a restricted 1 + N body problem

We started our studies with a planar Eulerian restricted four-body problem (ERFBP) where three masses move in circular orbits such that their configuration is always collinear. The fourth mass is small and does not influence the motion of the three primaries. In our model we assume that one of the primaries has mass 1 and is located at the origin and two masses of size µ rotate around it uniformly. The problem was studied in [3], where it was shown that there exist noncollinear equilibria, which are Lyapunov stable for small values of µ. KAM theory is used to establish the stability of the equilibria. Our computations do not agree with those given in [3], although our conclusions are similar. The ERFBP is a special case of the 1 + N restricted body problem with N = 2. We are able to do the computations for any N and find that the stability results are very similar to those for N = 2. Since the 1 + N body configuration can be stable when N > 6, these results could be of more significance than for the case N = 2.     

Eulerian limit for 2D Navier-Stokes equation and damped/driven KdV equation as its model

We discuss the inviscid limits for the randomly forced 2D Navier-Stokes equation (NSE) and the damped/driven KdV equation. The former describes the space-periodic 2D turbulence in terms of a special class of solutions for the free Euler equation, and we view the latter as its model. We review and revise recent results on the inviscid limit for the perturbed KdV and use them to suggest a setup which could be used to make a next step in the study of the inviscid limit of 2D NSE. The proposed approach is based on an ergodic hypothesis for the flow of the 2D Euler equation on iso-integral surfaces. It invokes a Whitham equation for the 2D Navier-Stokes equation, written in terms of the ergodic measures.     

Hierarchical Low Rank Approximation of Likelihoods for Large Spatial Datasets

Datasets in the fields of climate and environment are often very large and irregularly spaced. To model such datasets, the widely used Gaussian process models in spatial statistics face tremendous challenges due to the prohibitive computational burden. Various approximation methods have been introduced to reduce the computational cost. However, most of them rely on unrealistic assumptions for the underlying process and retaining statistical efficiency remains an issue. We develop a new approximation scheme for maximum likelihood estimation. We show how the composite likelihood method can be adapted to provide different types of hierarchical low rank approximations that are both computationally and statistically efficient. The improvement of the proposed method is explored theoretically; the performance is investigated by numerical and simulation studies; and the practicality is illustrated through applying our methods to two million measurements of soil moisture in the area of the Mississippi River basin, which facilitates a better understanding of the climate variability. Supplementary material for this article is available online.