We discuss dynamic server control in a two-class service system under a constraint on the number of high-priority customers. A class of randomized threshold policies is defined, and is proven to contain an optimal policy in the case without abandonments. The proof of optimality is then used to construct heuristic policies for the case of low-priority abandonments, which we test numerically. The...

We consider a memoryless single server queue with two classes of customers, customer types are distinguished by the payments and service discrimination (i.e. customers of the two classes are served with different service rates). Based on the different fixed payments of the two classes, we derive the priority distribution schemes from the perspective of profit maximization and social optimizatio...

This study investigates a critical issue for E-retailers: balancing their spending on information technology and advertising. E-retailers often over-spend on advertisement to attract customers, but fail to support them due to inadequate spending on technology infrastructure. In this study we show that the demand generation and site capacity must be synchronized. E-retailers should place a highe...

ABSTRACT:In this paper, we study a Markovian queueing model with balking for buffer sharing in ATM. It is assumed that server provides service with priorities termed as first, second and third. As soon as system accumulates L+1, M+1 customers third and second priority customers balk with balking function bn. We have obtained the steady state probabilities for the system by recursive method. Som...

In this paper we present a comparison of the service disciplines in real-time queueing systems (the customers have a deadline before which they should enter the service booth). We state that the more a service discipline gives priority to customers having an early deadline, the least the average stationary lateness is. We show this result by comparing adequate random vectors with the Schur-Conv...

The purpose of this chapter is to introduce the reader to the topic of strategic behavior of customers in queues. We assume that customers individually decide on issues such as whether or not to join the queue, whether or not to renege from a queue after waiting for a while, or whether or not to pay some fee in order to belong to a higher priority class. For some examples, we define the appropr...

This paper investigates a polling system with a random polling scheme, a 1-limited service discipline and deterministic service requirement modeling WLANs with QoS differentation capability. The system contains high and low priority queues that are distinguished via the probability of being served next. We propose a new iteration algorithm to approximate the waiting time of customers in the hig...

We consider a discrete-time cyclic-service system consisting of multiple stations visited by a single server. Customers from several priority classes arrive at an individual station according to independent batch Bernoulli processes. We assume a non-preemptive priority rule and non-zero switch-over times of the server between consecutive stations. 'Ve derive an exact expression for a weighted s...

The decision on the specification of a product development process based on the general platform is determined by the customer requirement/needs analysis, that is, the quality function deployment, through which common parts/components, options, and so on, are determined and MC level (number of derivative products) is decided through the requirements of customers. However, the current QFD method...