This paper investigates how AI-enabled dynamic pricing affects revenue, profitability and competitive dynamics in licenced sport merchandise retail. Although dynamic pricing is well studied in sport ticketing, its AI-enabled extension to licenced merchandise remains largely unexamined, and no prior sport management study has modelled it within a multi-channel retail ecosystem. The study addresses this gap by comparing three pricing approaches across online and offline channels: two transparent heuristic rules (rule-based and demand-responsive pricing) and a personalised algorithm that learns segment-level demand online, isolating what the AI-enabled learning layer adds over simpler automation.
An agent-based simulation model is developed with four agent types: online retailers, offline retailers, heterogeneous fan consumers and a sport organisation acting as licensor. The model integrates seasonal demand cycles, stochastic event shocks, an explicit word-of-mouth rule and a lost-sales inventory process. Rule-based pricing (RBP) and demand-responsive pricing (DRP) are specified as transparent heuristic pricing rules, whereas personalised dynamic pricing (PDP) adds an online logistic demand learner that updates segment-by-product elasticity beliefs and chooses expected-profit-maximising prices. Nine strategy combinations are compared over three simulated seasons, with one-at-a-time sensitivity analysis across seven uncertain parameters.
Under the specified model rules, algorithmic pricing configurations generate up to 26.4% higher three-season ecosystem revenue than the symmetric RBP baseline. The highest-revenue configuration is DRP online with RBP offline, while symmetric PDP produces the highest total profit. Offline retailers capture higher margins than online retailers when using DRP or PDP, and a counterintuitive result shows that simple online pricing can protect margins against algorithmically adaptive offline competitors. The wholesale cost ratio remains the most influential parameter for profitability.
This study is the first to model AI-enabled dynamic pricing for licenced sport merchandise within a multi-channel retail ecosystem, and the first to apply agent-based modelling to sport merchandise pricing. By treating the ecosystem as a complex adaptive system, it extends the sport pricing literature beyond ticketing and reveals emergent, ecosystem-level dynamics that conventional single-agent or equilibrium frameworks cannot capture.
