Telekom ağlarında kademeli fiyatlandırmayla kapasite kiralanması ve iş dağılımı.

Usage of data networks encompasses not only the traditional data applications but also newer applications such as realtime audio/video streaming, voice over TCP/IP, realtime transactions, asynchronous messaging and other batch transactions over digital networks. Each application has different capacity and quality of service (QoS) requirements. Each is affected differently by network reliability and speed. Major network providers have already started efforts to accommodate the QoS demand generated by these applications. Providers charge differently for the capacity they sell. For example, Internet service providers (ISPs) offer a combination of fixed price for a fixed maximum bandwidth (allyoucan send) or pay per hour (or minute) for again a maximum bandwidth. Wireless phone companies charge for text messaging based on bytes sent. Certain calling plans offer different per minute charges for phone conversations depending on when the call is placed. A firm uses data networks to perform and support business operations, which we will call tasks. For example, a video conference is a type of task, so is a remote web site update. The obvious commonality is that both tasks require network resources. A network resource is characterized by its capacity (bandwidth and duration) and QoS. We assume that the contract signed between the firm and the provider specifies the amount of bandwidth and quality guaranteed at a given time. Duration specifies the period during which the resource is available. There are two types of costs associated with using data networks. The first one is the resource (i.e., bandwidth, capacity) acquisition cost. The second is the opportunity cost incurred due to insufficient quality realized in performing certain tasks such as video conferencing. Given a heterogeneous set of tasks and resources that differ in bandwidth and quality requirements, it might be in the firm's best interest to use either multiple providers or sign multiple contracts with different bandwidth and quality requirements. For example, while resources with lower quality can be used for data applications, more expensive resources might be utilized for realtime applications with high QoS requirements. Therefore, the problem that the firm has to solve is a cost minimization problem that reflects a trade off between the cost of acquiring resources and the opportunity cost of degradation in realized quality of tasks performed. The quality of service of the resource affects the customer in two ways. First, size-fixed tasks might be delayed beyond acceptable deadlines. Second, the realized quality of a time-fixed task such as a video conference might be unacceptable creating an opportunity cost for the customer. In general, opportunity cost reflects the importance of a task. The more important the task is, the higher the penalty for not achieving desired targets (such as audio or visual quality). The decision maker will minimize the total cost by considering the tradeoff among these costs when assigning tasks to resources. In our study we formulate a cost minimization problem subject to QoS and capacity constraints. We consider the taxband pricing scheme suggested by Courcoubetis et al. (2000a) in which each supplier has a convex piece wise linear cost function for each resource offered. Courcoubetis et al. (2000a) claim that taxband pricing reduces bursty traffic since customers are likely to reduce such demand to avoid paying more. Suppliers in general would prefer having more customers with less capacity demand rather than fewer customers with high capacity demand. Given this pricing structure the customer has to decide how much capacity to acquire and how to allocate tasks. For tractability we relax the due date constraints and assume that all tasks and resources are available at time zero. We also assume that realtime applications have desired transmission rates and any deviation from that creates an opportunity cost for the customer. In spirit, this is the same objective function used in Kasap et al. (2007) that trades the quality cost with the capacity cost but the cost structure for capacity is quite different in this problem. In this study, firstly, we present taxband pricing. Secondly, a formulation and a solution procedure based on GBD are described. [ABSTRACT FROM AUTHOR]