Optimal control of parallel server systems with many servers in heavy traffic

We consider a parallel server system that consists of several customer classes and server pools in parallel. We propose a simple robust control policy to minimize the total linear holding and reneging costs. We show that this policy is asymptotically optimal under the many-server heavy traffic regime for parallel server systems when the service times are only server pool dependent and exponentially distributed. J.G. Dai’s research supported in part by National Science Foundation grants CMMI-0727400 and CNS-0718701, and by an IBM Faculty Award.     

Optimal birth control of free horizon problems for predator-prey system with age-structure

We study optimal birth policies for three age-dependent populations in a predator-prey system, which is controlled by fertility. New results on problems with free final time and integral phase constraints are presented, the approximate controllability of system is discussed.     

Optimal ℋ∞-state feedback control for continuous-time linear systemsfeedback control for continuous-time linear systems

This paper proposes a convex programming method to achieve optimal -state feedback control for continuous-time linear systems. State space conditions, formulated in an appropriate parameter space, define a convex set containing all the stabilizing control gains that guarantee an upper bound on the -norm of the closed-loop transfer function. An optimization problem is then proposed, in order to minimize this upper bound over the previous convex set, furnishing the optimal -control gain as its optimal solution. A limiting bound for the optimum -norm can easily be calculated, and the proposed method will achieve minimum attenuation whenever a feasible state feedback controller exists. Generalizations to decentralized and output feedback control are also investigated. Numerical examples illustrate the theory.This research has been supported in part by grants from Fundação de Amparo à Pesquisa do Estado de São Paulo—FAPESP and Conselho Nacional de Desenvolvimento Científico e Tecnológico, CNPq—Brazil. The authors are grateful to the anonymous referees for their useful comments on this paper.     

Optimal boundary control of the wave equation with pointwise control constraints

In optimal control problems frequently pointwise control constraints appear. We consider a finite string that is fixed at one end and controlled via Dirichlet conditions at the other end with a given upper bound M for the L ^∞-norm of the control. The problem is to control the string to the zero state in a given finite time. If M is too small, no feasible control exists. If M is large enough, the optimal control problem to find an admissible control with minimal L ²-norm has a solution that we present in this paper.     

Optimal variational principle for backward stochastic control systems associated with Lévy processes

The paper is concerned with optimal control of backward stochastic differential equation (BSDE) driven by Teugel’s martingales and an independent multi-dimensional Brownian motion,where Teugel’s martin- gales are a family of pairwise strongly orthonormal martingales associated with Lévy processes (see e.g.,Nualart and Schoutens’ paper in 2000).We derive the necessary and sufficient conditions for the existence of the op- timal control by means of convex variation methods and duality techniques.As an application,the optimal control problem of linear backward stochastic differential equation with a quadratic cost criteria (or backward linear-quadratic problem,or BLQ problem for short) is discussed and characterized by a stochastic Hamilton system.     

Parallel difference schemes with interface extrapolation terms for quasi-linear parabolic systems

In this paper some new parallel difference schemes with interface extrapolation terms for a quasi-linear parabolic system of equations are constructed. Two types of time extrapolations are proposed to give the interface values on the interface of sub-domains or the values adjacent to the interface points, so that the unconditional stable parallel schemes with the second accuracy are formed. Without assuming heuristically that the original boundary value problem has the unique smooth vector solution, the existence and uniqueness of the discrete vector solutions of the parallel difference schemes constructed are proved. Moreover the unconditional stability of the parallel difference schemes is justified in the sense of the continuous dependence of the discrete vector solution of the schemes on the discrete known data of the original problems in the discrete W2(2,1) (Q△) norms. Finally the convergence of the discrete vector solutions of the parallel difference schemes with interface extrapolation terms to the unique generalized solution of the original quasi-linear parabolic problem is proved. Numerical results are presented to show the good performance of the parallel schemes, including the unconditional stability, the second accuracy and the high parallelism.     

Optimal control for a population dynamics with age-dependent and diffusion in a periodic environment

Here we consider the optimal harvesting problem for a periodic age-dependent population dynamics with diffusion. Namely, we investigate the model with periodic vital rates and a periodic forcing term that sustains oscillations. By using Mazur’s theorem, we demonstrate existence of solutions of the optimal control problem and by the conception of normal cone, we also obtain the first order necessary conditions of optimality for the problem. Our results extend some known criteria.     

Optimal distributed control of nonlinear Cahn–Hilliard systems with computational realization

The goal of this work is to present some optimal control aspects of distributed systems described by nonlinear Cahn–Hilliard equations (CH). Theoretical conclusions on distributed control of CH system associated with quadratic criteria are obtained by the variational theory (see [17]). A computational result is stated by a new semi-discrete algorithm, constructed on the basis of the finite-element method with the updated (nonlinear) conjugate gradient method for minimizing the performance index efficiently. Finally, the implementation of a laboratory demonstration is included to show the efficiency of the proposed nonlinear scheme.     

A control point based curve with two exponential shape parameters

A generalization of a recently developed trigonometric Bézier curve is presented in this paper. The set of original basis functions are generalized also for non-trigonometric functions, and essential properties, such as linear independence, nonnegativity and partition of unity are proved. The new curve—contrary to the original one—can be defined by arbitrary number of control points meanwhile it preserves the properties of the original curve.     

Nonlinear stochastic perturbations of dynamical systems and quasi-linear parabolic PDE’s with a small parameter

In this paper, we describe the asymptotic behavior, in the exponential time scale, of solutions to quasi-linear parabolic equations with a small parameter at the second order term and the long time behavior of corresponding diffusion processes. In particular, we discuss the exit problem and metastability for the processes corresponding to quasi-linear initial-boundary value problems.     

Admission control for a multi-server queue with abandonment

In a M/M/N+M queue, when there are many customers waiting, it may be preferable to reject a new arrival rather than risk that arrival later abandoning without receiving service. On the other hand, rejecting new arrivals increases the percentage of time servers are idle, which also may not be desirable. We address these trade-offs by considering an admission control problem for a M/M/N+M queue when there are costs associated with customer abandonment, server idleness, and turning away customers. First, we formulate the relevant Markov decision process (MDP), show that the optimal policy is of threshold form, and provide a simple and efficient iterative algorithm that does not presuppose a bounded state space to compute the minimum infinite horizon expected average cost and associated threshold level. Under certain conditions we can guarantee that the algorithm provides an exact optimal solution when it stops; otherwise, the algorithm stops when a provided bound on the optimality gap is reached. Next, we solve the approximating diffusion control problem (DCP) that arises in the Halfin–Whitt many-server limit regime. This allows us to establish that the parameter space has a sharp division. Specifically, there is an optimal solution with a finite threshold level when the cost of an abandonment exceeds the cost of rejecting a customer; otherwise, there is an optimal solution that exercises no control. This analysis also yields a convenient analytic expression for the infinite horizon expected average cost as a function of the threshold level. Finally, we propose a policy for the original system that is based on the DCP solution, and show that this policy is asymptotically optimal. Our extensive numerical study shows that the control that arises from solving the DCP achieves a very similar cost to the control that arises from solving the MDP, even when the number of servers is small.     

Some aspects of reachability for parabolic boundary control problems with control constraints

A class of one-dimensional parabolic optimal boundary control problems is considered. The discussion includes Neumann, Robin, and Dirichlet boundary conditions. The reachability of a given target state in final time is discussed under box constraints on the control. As a mathematical tool, related exponential moment problems are investigated. Moreover, based on a detailed study of the adjoint state, a technique is presented to find the location and the number of the switching points of optimal bang-bang controls. Numerical examples illustrate this procedure.     

Special features of strongly coupled systems of convection–diffusion equations with two small parameters

Strong coupling of convection–diffusion equations with two small parameters generates a solution decomposition which differs significantly from that for the one-parameter case. We explain the basic features and prove pointwise estimates for the first-order derivatives which allow us to analyze the upwind finite difference scheme on layer-adapted meshes.     

Optimal admission and pricing control problem with deterministic service times and sideline profit

We consider a discrete-time admission control problem in a company operating in service industries with two classes of customers. For the first class of customers, the company then (1) has an option to accept or reject him/her (admission control), or (2) decides on an offering price (pricing control). The second-class (sideline) customers are only served if no first-class customers are in the system, and this yields the sideline profit. In this paper, we discuss both admission control and pricing control problems within an identical framework, and we examine the properties of the optimal policies to maximize the total expected present discounted net profits. We show that when the sideline profit is large, the optimal policies may not be monotone in the number of first-class customers in the system.     

Delay-range-dependent synchronization criterion for Lur’e systems with delay feedback control

In this paper, a master–slave synchronization scheme is investigated by using feedback control mechanism with time-varying delay. The time-delay is assumed to be a time-varying continuous function belonging to a given range. By constructing a novel Lyapunov–Krasovskii functional, which includes the information of the range, new delay-range-dependent synchronization criterion is established in term of LMI. It is shown that the new criterion improve some of the previous results in the earlier references. Simulation example is given to show the effectiveness and less conservatism of the proposed criterion.     

Optimal proportional reinsurance model with transaction costs

Based on the mixed control strategy (regular control and impulse dividend control strategy), we formulate a proportional reinsurance model with transaction costs. For getting the maximal return function and associated mixed control strategy, using Itô calculus and classical mixed control theory, we derive the quasi-variational inequality solution to this optimal problem. Furthermore, we obtain its closed forms under some assumptions.