Ning Nan

Although contact tracing apps are effective in containing the impacts of COVID-19, their centralized management model cause significant user concerns about privacy and information security. Blockchain-based contact tracing apps have received increasing attention as a better alternative. However, the adoption of blockchain-based solutions is slow. This paper investigates how user perceptions and technical features of blockchain-based contact tracing apps individually and jointly influence users' adoption intention in the short and long term. The collection of apps and users is viewed as a Complex Adaptive System. An Agent-Based Modeling technique is adopted in this paper, where users are regarded as agents. In addition, agents are adaptive to the environment and interact with each other according to behavioral rules. Via a trust-centered lens, several simulation studies were conducted with different pre-determined parameters to investigate the impacts of various characteristics of users and solutions. Finally, survey-based validations and sensitivity analyses were conducted to validate the rationality of the model. The results show that both cognitive and emotional trusts of the users are positive and crucial to the adoption diffusion. Regarding app characteristics, compatibility level and scalability resistance could positively affect adoption. Moreover, service providers can choose to invest in highly connected communities. This study takes the first step to developing a dynamic diffusion model of blockchain-based COVID-19 contact tracing apps. In addition, it provides theoretical and practical implications to facilitate the implementation of blockchain-based solutions.

View record

Blockchain, or distributed ledger technology (DLT), is expected to enable a highly decentralized and trusted business environment. Yet the business pursuit for profit maximization calls for a more centralized structure and thereby conflicts with the decentralized ideology of blockchain. In the blockchain-driven supply chain finance (SCF) platforms, focal buyer companies can issue “cash tokens” which are blockchain-based electronic invoices/certificates with due days of payment and pay the “cash tokens” to their direct suppliers instead of using traditional open-account techniques. Direct suppliers can hold the “cash tokens” to maturity, use the “cash tokens” as collateral for loans, or pay the “cash tokens” to the upper tier suppliers in goods transactions (we call the last option “token delivery” in the following). Because the “cash tokens” are delivered tier by tier based on transactions, mid-tier suppliers can become a “bottleneck” in the blockchain-driven SCF application. In this paper, we consider the supply chain network as a complex system where firms are self-organized and adaptive to their competitive environment. Via this theoretical lens, we investigate how the mid-tier suppliers’ token delivery and supply chain transaction structures interplay over time in the blockchain-driven SCF platforms; meanwhile, how industry characteristics such as the number of firms in each tier and firm size in each tier can influence the interactions. We propose that in the short term, blockchain technology increases mid-tier suppliers’ transaction efficiency and thus motivates mid-tier suppliers’ token delivery and promotes the decentralization of supply chain transaction structure, i.e., upper-tier suppliers make new transaction links with mid-tier suppliers and focal buyers; in the long term, the more decentralized supply chain transaction structure will in turn negatively affect mid-tier suppliers’ token delivery motivation and drive the supply chains more centralized, i.e., the upper-tier suppiers start to face financing difficulties again and some of the transaction links will diminish. Besides, supply chains with a flatter organizational structure, a larger setup cost gap between mid-tier suppliers and upper-tier suppliers, and a higher operation cost of upper-tier suppliers can remain decentralized longer. We will test our theoretical propositions by a series of simulation experiments in an agent-based model.

View record