Modelowanie agentowe dyfuzji samochodów elektrycznych
[Agent-based modeling of the diffusion of electrical vehicles]

Karolina Bienias (KBO, PWr, Wrocław)

 
International agreements on fighting climate change require countries to lower 
their use of fossil fuels. One way of doing this is to increase the proportion 
of vehicles powered by electricity. Using an agent-based (AB) modeling 
approach, we study the temporal dynamics of consumers' (agents') opinions 
regarding electrical cars and decisions to purchase them. In practice, the 
adoption of electrical cars is driven by: 
(a) the individual situation of agent (e.g. income, the number of cars a 
household possesses and the distance which is covered by agent daily) 
(b) local influences that come from interactions between agents (e.g. word of 
mouth, agents comparing their opinion with their neighbors) and 
(c) the external field (e.g. advertising, prices, subsidies and the possibility 
of charging in car parks). 

We model this process within an urban population using an AB model developed in 
the NetLogo environment. The way in which we define and model agents' immunity 
to (or dependence on) social influence distinguishes our model from other 
approaches. In our model, at each moment of discrete time, we consider agents 
who are ready to buy a new car. The decision to purchase an electrical vehicle 
(EV) is based on an individual's income exceeding an individual threshold, 
which depends on the difference in the costs of conventional vehicles and EVs, 
the attitudes and habits of agents, advertising and the infrastructure 
developed. Using Monte Carlo simulations we show how the purchase of electrical 
cars diffuses across society. On the basis of these results, we make some 
policy recommendations for governments wishing to promote the use of electrical 
vehicles.