A service center is a facility with multiple heterogeneous servers providing specialized service to multiple types of customers. An assignment policy specifies which server is enabled to serve which types of customer, and a routing policy specifies which server a customer will be routed to for service. Thus, a server can be enabled to serve many types of customer, and a customer may have many alternate servers who can serve him. This paper aims to provide decision models to determine optimal static assignment and routing policies, explicitly taking into account the stochastic fluctuations of demand along with the autocorrelations and cross-correlations of the different traffic streams. We consider several possible performance measures and formulate the optimization problem as a mixed integer nonlinear programming problem. We also develop an efficient heuristic algorithm to enhance scalability. Finally, we compare the different policies using the heuristic algorithms. We observe numerically that the routing policy tries to combine the negatively correlated traffic streams, and separate the positively correlated traffic streams.

This work focuses on Exchangeable Occupancy Models (EOMs) and their relations with the Uniform Order Statistics Property (UOSP) for point processes in discrete time. As our main purpose, we show how definitions and results presented in Shaked, Spizzichino, and Suter [8] can be unified and generalized in the frame of occupancy models. We first show some general facts about EOMs. Then we introduce a class of EOMs, called ℳ(a)-models, and a concept of generalized Uniform Order Statistics Property in discrete time. For processes with this property, we prove a general characterization result in terms of ℳ(a)-models. Our interest is also focused on properties of closure w.r.t. some natural transformations of EOMs.

In this paper, we propose a stochastic method to project the public pension fund in the public pension system (PPS). For this we introduce the stochastic differential equations for the three parts: the premium revenue, the benefit expenditure, and the fund process. From these we show that the solution of the aggregated fund process is the sum of log-normals, which is approximated as one log-normal for the analytic result. Related to the parameter estimations, we implement the moment matching in the first moment. For the second moment, we apply the extreme value method following Parkinson. In order to follow Parkinson, we take the maximum and the minimum range of the fund amount based on the various sensitivity result as well as the baseline one from the deterministic projection result. In this reason, it is naturally to maintain the close interrelation with the deterministic projection result, which is very important since it is still key result in the actuarial valuation of the PPS.

Two simple approximation formulas for approximating the areas under the t distribution are presented here which are easily programmable on pocket calculators. Their accuracy should be sufficient for most of the practical applications.

We consider a semi-Markov additive process A(·)—that is, a Markov additive process for which the sojourn times in the various states have general (rather than exponential) distributions. Letting the Lévy processes Xi(·), which describe the evolution of A(·) while the background process is in state i, be increasing, it is shown how double transforms of the type can be computed. It turns out that these follow, for given nonnegative α and q, from a system of linear equations, which has a unique positive solution. Several extensions are considered as well.

This paper is devoted to the study of the asset allocation problem for a DC pension plan with minimum guarantee constraint in a hidden Markov regime-switching economy. Suppose that four types of assets are available in the financial market: a risk-free asset, a zero-coupon bond, an inflation-indexed bond and a stock. The expected return rate of the stock depends on unobservable economic states, and the change of states is described by a hidden Markov chain. In addition, the CIR process is used to describe the evolution of the nominal interest rate. The contribution rate is also assumed to be stochastic. The goal of investment management is to minimize the convex risk measure of the terminal wealth in excess of the minimum guarantee constraint. First, we transform the partially observable optimization problem into the one with complete information using the Wonham filtering technique and deal with the minimum guarantee constraint by constructing auxiliary processes. Furthermore, we derive the optimal investment strategy by the BSDE approach. Finally, some numerical results are presented to illustrate the impacts of some important parameters on investment behaviors.

We consider a process that starts at height y, stays there for a time X0 ∼ exp(y) when it drops to a level Z1 ∼ U(0, y). Thereafter it stays at level Zn for time Xn ∼ exp(Zn), then drops to a level Zn+1 ∼ U(0,Zn). We investigate properties of this process, as well as the Poisson hyperbolic process which is obtained by randomizing the starting point y of the above process. This process is associated with a rate 1 Poisson process in the positive quadrant: Its path is the minimal RCLL decreasing step function through Poisson points in the positive quadrant. The finite dimensional distributions are then multivariate exponential in sense of Marshall-Olkin.

Single-server queueing systems with mixed traffic are a flexible modeling tool that have been used to analyze polling systems and database striping strategies. They also have applications in manufacturing and telecommunication systems. Ordinary (open) customers arrive according to a Poisson process, receive service, and leave the system. Permanent (closed) customers remain in the system, reentering the queue after receiving service. This work examines the transient behavior of the permanent customers. The cycle times and output/departure process for the permanent customers are described.

We consider the problem of scheduling jobs on parallel machines to minimize flowtime, where all decisions about loading and sequencing must be made before any processing is done. We find policies that minimize flowtime stochastically or in the increasing convex sense under various assumptions.

In this article we study Markov-modulated queues with and without jumps. These queuing models arise naturally in production-inventory systems with and without an external supplier. We show an interesting decomposition property that relates the equilibrium state distributions in these two systems and present an integrated warranty-inventory management model as an application.

Several agents with different subjective probabilities make a binary decision at a time determined by a planner. Each agent chooses the action that has the highest probability of success. Given that their probabilities differ, so will their choices. From time 0 until decision time, all the agents are entitled to access the same increasing flow of information. The planner, who gains from having as many agents as possible making the right choice, faces the following tradeoff: the more information she feeds to the agents, the better off they will be in making their decisions, but the less likely they will be to diversify their actions, so the more difficult it will be for her to hedge her positions. The model gives rise to a continuous time optimal stopping problem.

Consider a system consisting of a number of modules. The probability that a particular module is among the ones that have already failed by the time of system failure can be used as a measure of the importance of that module. In El-Neweihi and Sethuraman [6], this probability was called the role of a module in the failure of a system and its properties were developed. In this article we propose the number of failed modules from among a particular subgroup of modules as the basis for measuring the role of that subgroup of modules. We study some monotonicity properties of the distribution of that random variable, in some second order r-out-of-k systems. We illustrate the possible extensions of these results to multistate systems. Yet another measure of the role of a particular subgroup of modules can be based on the number of failed components from this subgroup by the time of system failure. We derive some monotonicity properties of the expected value of the number of such failed components.

In 1987, Sheldon M. Ross [4] gave an interesting recurrent relation method for approximations of important functions in renewal theory. Using his idea and Feller's technique for inverting Laplace transforms, we obtain a general method for recurrent approximations to the solutions of renewal-type integral equations.

Recent advances in the mathematical analysis of flow control have prompted the creation of the Scalable TCP (STCP) and Exponential RED (E-RED) algorithms. These are designed to be scalable under the popular deterministic delay stability modeling framework. In this article, we analyze stochastic models of STCP and STCP combined with E-RED link behavior. We find that under certain plausible network conditions, these probabilistic models also exhibit scalable behavior. In particular, we derive parameter choice schemes for which the equilibrium coefficients of variation of flow rates are bounded, however large, fast, or complex the network. Our model is shown to exhibit behavior similar to the mean field convergence that has recently been observed in TCP.

There is a set of n bandits and at every stage, two of the bandits are chosen to play a game, with the result of a game being learned. In the “weak regret problem,” we suppose there is a “best” bandit that wins each game it plays with probability at least p > 1/2, with the value of p being unknown. The objective is to choose bandits to maximize the number of times that one of the competitors is the best bandit. In the “strong regret problem”, we suppose that bandit i has unknown value vi, i = 1, …, n, and that i beats j with probability vi/(vi + vj). One version of strong regret is interested in maximizing the number of times that the contest is between the players with the two largest values. Another version supposes that at any stage, rather than choosing two arms to play a game, the decision maker can declare that a particular arm is the best, with the objective of maximizing the number of stages in which the arm with the largest value is declared to be the best. In the weak regret problem, we propose a policy and obtain an analytic bound on the expected number of stages over an infinite time frame that the best arm is not one of the competitors when this policy is employed. In the strong regret problem, we propose a Thompson sampling type algorithm and empirically compare its performance with others in the literature.

Binary series-parallel (BSP) graphs have applications in transportation modeling, and exhibit interesting combinatorial properties. This work is limited to the second aspect. The BSP graphs of a given size – measured in edges – are enumerated. Under a uniform probability model, we investigate the asymptotic distribution of the order (number of vertices) and the asymptotic average length of a random walk (under different strategies) of large graphs of the same size. The systematic method throughout is to define the graphs, and the features we evaluate by a structural equation (equivalent to a regular expression). The structural equation is translated into an equation for a generating function, which is then analyzed.

We study a cumulative storage system that is totally cleared sporadically at stationary renewal times and whenever a finite-capacity threshold is exceeded. The independent and identically distributed inputs occur at time epochs that also form a stationary renewal process. We determine the distribution of the interoverflow times. Although this distribution is quite intricate when both underlying renewal processes are general, in the special case of Poisson sporadic clearings we obtain a neat formula for its Laplace transform.

We consider a system consisting of a single server serving a fixed number of stations. At each station, there is an infinite queue of customers that have to undergo a preparation phase before being served. This model is connected to layered queueing networks, to an extension of polling systems and surprisingly to random graphs. We are interested in the waiting time of the server. For the case where the server polls the stations cyclically, we give a sufficient condition for the existence of a limiting waiting-time distribution and we study the tail behavior of the stationary waiting time. Furthermore, assuming that preparation times are exponentially distributed, we describe in depth the resulting Markov chain. We also investigate a model variation where the server does not necessarily poll the stations in a cyclic order, but always serves the customer with the earliest completed preparation phase. We show that the mean waiting time under this dynamic allocation never exceeds that of the cyclic case, but that the waiting-time distributions corresponding to both cases are not necessarily stochastically ordered. Finally, we provide extensive numerical results investigating and comparing the effect of the system's parameters to the performance of the server for both models.

We develop a new measure of reliability for the mean behavior of a process by calculating the probability the cumulative sample mean will stay within a given distance from the true mean over a period of time. This probability is derived using boundary-crossing properties of Brownian bridges. We derive finite sample results for independent and identically distributed normal data, limiting results for data meeting a functional central limit theorem, and draw parallels to standard normal confidence intervals. We deliver numerical results for i.i.d., dependent, and queueing processes.