Modeling the diffusion of residential photovoltaic systems in Italy: An agent-based simulation

We propose an agent-based model to simulate how changes to the Italian support scheme will affect the diffusion of PV systems among single- or two-family homes. The adoption decision is assumed to be influenced by (1) the payback period of the investment, (2) its environmental benefit, (3) the household's income, and (4) communication with other agents. The estimation of the payback period considers investment costs, local irradiation levels, governmental support, earnings from using self-produced electricity vs. buying electricity from the grid, administrative fees, and maintenance costs. The environmental benefit is estimated by a proxy for the CO2 emissions saved. The household income accounts for the specific economic conditions across different regions and the agent's age group, level of education, and household type. Finally, the influence of communication is measured by the number of links with other households that have already adopted a PV system. In each simulation step, the program dynamically updates the social system and the communication network, while the PV system's investment costs are revised according to a one-factor experience curve. The model's structure is applied for a case study based on the evolution of residential PV systems in Italy over the 2012-2026 period. The model's initial state is calibrated on the basis of the actual diffusion of residential PV in Italy over the 2006-2011 period. Our results show that, following Italy's new feed-in tariff scheme, domestic PV installations are already beyond an initial stage of rapid growth and, though likely to spread further, they will do so at a significantly slower rate of diffusion.