Simulation-based models for B2B platform dynamics

Abstract

Business-to-business (B2B) trade has historically been bilateral: the complexity and specificity of industrial products make it difficult to standardize transactions, limiting the emergence of platform- mediated exchange. Recent advances in artificial intelligence driven automation are changing this. Platforms can now automate supplier discovery, qualification, negotiation, and contracting at scale, making platform-mediated complex sourcing increasingly feasible. As adoption grows, network effects become plausible in B2B markets, raising the possibility of critical mass dynamics and potentially monopolistic concentration analogous to what has occurred in business-to-consumer platforms. This paper reviews simulation-based studies of B2B platform diffusion across six structural dimensions: methodology, actor representation, network topology, platform scope, adoption dynamics, and empirical calibration. Agent-based modeling (ABM) offers the necessary actor-level resolution, yet within the reviewed sample no study combines heterogeneous firms, structured supply-chain ties, and procurement- specific adoption rules based on trading-partner presence. Multi-platform competition was also not identified in the reviewed sample. On this basis, the paper outlines design requirements for a future ABM in which diffusion emerges from supply- chain topology, without implementing the model itself, as a useful next step toward understanding B2B platform adoption dynamics.