Integrating scenario analyses and risk preferences into carbon-regulated supply chains: a dynamic repeated game-theoretic optimisation approach

To remain competitive while meeting sustainability goals, manufacturers' supply chain models subject to carbon regulation and serving eco-conscious markets need to adjust their production and abatement choices dynamically. However, existing models ignore how endogenous risk preferences and exogenous scenario uncertainty combine to drive these strategic shifts. This paper integrates Scenario Analyses and Risk Preferences–risk-averse, risk-neutral, and risk-seeking behaviours under optimistic, neutral, and pessimistic market scenarios–into a Repeated Carbon Game-theoretic (SA-RP-RCG) framework for a carbon-regulated supply chain model. This model applies repeated-game logic to decompose long-term dynamics into a series of non-convex Mixed-Integer Quadratically Constrained Programming (MIQCP) sub-problems, each of which performs an optimisation trade-off between profitability and emissions. Numerical results from the e-scooter industry show that production increases only when subsidies accompany moderate carbon taxes, thereby widening profit gaps in favour of carbon-efficient industry leaders, yet allowing mid-range manufacturers to narrow the gap if they invest early. Moderate risk tolerance yields stable performance, whereas excessive risk-seeking reduces returns for carbon-efficient manufacturers. Less carbon-efficient manufacturers can improve their position by adopting bolder strategies, such as risk-seeking. These findings help asymmetric manufacturers align their decisions with evolving market signals and policy interventions while selecting an appropriate risk posture.