Q-Safe: QoS-Aware Pricing Scheme for Provisioning Safety-as-a-Service

In this paper, we propose a Quality of Service (QoS)-aware pricing scheme, termed as Q-Safe, for provisioning safety-related decisions to the end-users. Typically, a Safe-aaS platform provides customized decisions to the end-users, as per the latter's requirement. During the registration of the end-users, they provide their source and destination locations, select certain decision parameters, and make payment through the Web portal. Based on these selected decision parameters, the decision is generated. In the proposed pricing scheme, we consider the presence of multiple Safety Service Providers (SSPs) in the Safe-aaS platform. Therefore, the end-users possess the opportunity to select a SSP, depending on the price charged by the latter. On the other hand, the end-users may compromise with the quality of the decision provided through the selection of the available safety services at a low cost. Considering road transportation as the application scenario of Safe-aaS and to address these above-mentioned issues, we propose a dynamic pricing scheme, Q-Safe. We introduce the concept of varying prices to be charged by the SSPs for each of the decision parameters, based on the fluctuation in the value of these parameters with time. Each of the end-users selects certain decision parameters among the ones displayed in the Web portal. Thereafter, the SSPs suggest decision parameters to the end-users depending upon their present geographical location. To model these interactions between the SSPs and the end-users, we map the scenario with Non-Cooperative
Multiple Leader Multiple Follower Stackelberg
game, where the SSPs act as leaders and the end-users act as followers. Exhaustive analysis of our proposed scheme demonstrates that the average profit of the SSP is improved by 70.88%, 52%, and 77% compared to the Per-Subscriber model (Guijarro
et al.
2016), PRIME (Roy
et al.
2020), and RegPrice (Roy
et al.
2021) in the presence of 200 sensor nodes in the simulation environment. Additionally, we characterize the errors during the estimation of energy consumed, utility, effective time, and total cost, with the increase in the number of end-users.