Predictive Control Strategies for Sustaining Innovation Adoption on Multilayer Social Networks
Martina Alutto, Qiulin Xu, Fabrizio Dabbene, Hideaki Ishii, Chiara Ravazzi
TL;DR
The paper addresses sustaining diffusion of an innovation in a population where adoption interacts with evolving opinions on a multilayer network. It develops a coupled SIV-style adoption model with Friedkin–Johnsen opinion dynamics, analyzes adoption-free and adoption-diffused equilibria, and derives stability conditions. A Model Predictive Control framework with three levers—opinion shaping, adoption-propensity enhancement, and dissatisfaction reduction—demonstrates that predictive, adaptive interventions can sustain higher long-term adoption at comparable or lower costs than constant policies. The results highlight a practical toolkit for policymakers to promote sustainable diffusion using scalable, forward-looking control strategies.
Abstract
This paper studies an optimal control problem for an adoption-opinion model that couples innovation adoption with opinion formation on a multilayer network. Adoption spreads through social contagion and perceived benefits, while opinions evolve via social interactions and feedback from adoption levels. Individuals may abandon adoption due to dissatisfaction or external constraints, potentially hindering diffusion. We analyze system equilibria and their stability, identifying conditions under which adoption persists. We introduce a Model Predictive Control (MPC) strategy that dynamically adapts interventions to the predicted system evolution. Three types of control are compared: shaping opinions, acting on the adoption rate, and reducing dissatisfaction. Overall, MPC interventions outperform static constant control, achieving higher adoption at comparable or lower cost. These results highlight the potential of predictive, adaptive strategies to support sustainable behavior diffusion, offering policymakers scalable tools for effective interventions.