Regulation-as-a-Service: Model Checking for Decision-Making Behaviors in Price-Sensitive Service Systems

Abstract

Gig economy is growing rapidly with the benefits of Internet progress. Loads of researches attempt to help gig service platforms design pricing strategies by diving into supply-demand modeling and market equilibrium technologies. In these studies, workers’ noncompliance behaviors, such as fatigued driving in ride-hailing services that may cause car accidents, are ignored. Thus, forming a pricing strategy that can avoid noncompliance behaviors is urgent and challenging. In this paper, we propose a price-based computing framework to regulate workers’ behaviors, called Regulation as a Service (RaaS). We abstract workers’ decision-making behaviors as Markov Decision Process (MDP), and exploit model checking with formal method to prove the effectiveness of proposed incentive mechanism, through which we can provide rational decision formulations for gig service platforms to make comprehensive pricing strategies. Moreover, RaaS can be flexibly incorporated into existing price-sensitive service systems. Experiments are conducted to illustrate the advanced nature and practicability of RaaS.