Considering the different information acquired by users in the existing and new service scenarios, we further explore the signaling effect of compensation on consumer quality perception. We investigate the interaction between compensation and price, as well as the effect of participants’ risk attitudes on the CSP’s optimal decisions. Based on mathematical game models, this study develops the CSP’s profit-maximizing compensation and pricing strategies for existing and new cloud service scenarios, respectively. Moreover, few studies have considered all possible risk attitudes of both the CSP and customers when guiding the CSP to make decisions. However, the research related to cloud services mainly has focused on the issue of pricing strategies, underestimating the role of compensation. All rights reservedĪppropriate pricing and compensation terms in service level agreements (SLAs) help cloud service providers (CSPs) attract users and make profits. Third, for partial‐refund policies, more often than not, social learning increases social welfare when the product quality is high specifically, in many cases, it increases not only the profit of the seller but also the welfare of consumers. Hence, sellers may finally switch to the partial‐refund policy. Second, under social learning, the partial‐refund policy tends to be more beneficial to sellers than both full‐refund and no‐refund policies although, when the product quality is high, the no‐refund policy tends to bring more benefits to sellers than the full‐refund policy. It will cause the no‐refund sellers to choose higher prices and inventory, and the partial‐refund sellers to set lower prices and refund amounts. First, when sellers have relatively higher expectation of product quality (or simply the product quality is high), social learning makes the sellers offering either no‐refund policies or partial‐refund policies better off in terms of the increased profit. In this study, we investigate how social learning interacts with sellers' return policies. However, with the development of online review platforms, an increasing number of consumers are engaging in social learning by referring to others' reviews to reduce valuation uncertainty. Sellers are conventionally generous with their return policies for valuation‐uncertain products, such as experience products and new products.
Third, the existence of PDSL increases the provider’s profit in all four market scenarios as long as the provider’s capacity is larger than users’ prior mean QoS belief. Second, users’ congestion awareness causes the provider to set a non‐decreasing pricing policy in the two‐period market, while the provider’s steady‐state pricing policy in the infinite‐period market increases with the capacity and the prior QoS belief. In a two‐period market with congestion‐unaware users, the provider would always set a flat price when there is no PDSL in contrast, when there is PDSL, the optimal price can either increase or decrease, depending on the capacity and the prior mean QoS belief. First, the presence of PDSL significantly affects the provider’s optimal pricing policy. Our analysis yields several key insights. We study how such a new learning process affects the service provider’s dynamic pricing strategy in four different market scenarios, depending on whether the decisions are for two periods or infinite periods and whether users are aware of the congestion effect or not. The key difference from the traditional social learning mechanism is that the learning object (QoS) is not a fixed value, but instead, it depends on the number of review participants. In this paper, we consider users’ participation dependent social learning (PDSL), i.e., learning of the QoS through participants’ online reviews. For a new online service, the potential users are often uncertain about both the capacity and congestion level of the service, and hence are uncertain about the quality of service (QoS). The quality of many online services (such as online games, video streaming, cloud services) depends not only on the service capacity but also on the number of users using the service simultaneously.
0 kommentar(er)
