Sustainable production pathways for the photovoltaic supply chain under circular economy considerations amid uncertain demand

The rapid expansion of the photovoltaic industry has brought about the dual challenges of structural supply-demand imbalances and the large-scale retirement of modules. However, existing research has struggled to effectively address the supply-demand matching issue between recycling and manufacturing within the photovoltaic closed-loop supply chain. To this end, this study constructs a multi-objective optimization model for the photovoltaic closed-loop supply chain based on module heterogeneity. The model employs a Gaussian process regression forecasting model and fuzzy chance-constrained programming to model demand uncertainty. The results indicate that the setting of the fuzzy confidence level for demand has a significant impact on production strategies. When the confidence level is set to 0.5, total production can reach 3000 GW by 2035, which aligns more closely with mainstream demand forecasts. In terms of technology routes, N-type modules will become the mainstream choice in the future market, with HJT modules showing particular development potential due to their advantages in lower material and energy consumption. From a long-term perspective, the recycling of PV modules helps enhance the overall profitability of the supply chain. Recycling price analysis further indicates that maintaining recycling prices within the range of 400 to 500 yuan/t can more effectively balance profit distribution among supply chain participants. This study provides decision-making support for photovoltaic enterprises in production planning under strategies that balance profits and carbon emissions.