Innovation diffusion models

-Wind Power
Wind Power technology is analyzed in terms of diffusion for the leading countries, with incentive effects introduced as exogenous dynamics in the Generalized Bass Model (GBM) framework. The GBM framework is also compared, in terms of forecast accuracy over a set of different accuracy measures and forecasting horizons, to the Standard Bass, Logistic, and Gompertz models. 

-Nuclear Energy
Nuclear Energy diffusion of some graduated developing countries (the Slovak Republic and South Korea) and some developing countries (Ukraine, China, Bulgaria, and India), is analyzed in a GBM framework. Considering 50 years as a reasonable lifetime for reactors, the depletion time of uranium has been estimated using both a GBM and OECD forecasts, with the uranium requirements scheduled for 2035.

For more details about the research in innovation diffusion models of the research group in Padua, see here.



-Pleural Mesothelioma
A model, based on cellular automata, was developed to predict the future evolution of the number of deaths due to Pleural Mesothelioma among residents in the area around Casale Monferrato (Italy). The model is adapted to consider the individual exposure history into account (i.e., the duration of the exposure and the pollution level by asbestos in that period).


Natural Phenomena

-Diffuse Solar Radiation
A new multiple regression model was developed to estimate hourly values of diffuse solar radiation at the surface in terms of global solar radiation at the surface, which took into account explicitly the effects of cloud, air pollution and seasonal variation of the climate. There is evidence that by including additional cloud information such as the cloudiness and cloud height, it is possible to develop a simple regression model that performs well for hourly values as well.

-Large Volcanic Eruptions
A poisson process is used to model the largest volcanic eruptions in earth, with an intensity function that takes into account the recording bias expecially for eruptions of lower magnitude. The model has been implemented in a Bayesian context and the intensity function is based on a change-point model.

-Coastal Erosion
In the Holderness Coast (UK), since 1951 a monitoring program has been started in 118 stations along the coast, providing an invaluable, but often missing, source of information. A hierarchical random effect models, in a Bayesian context, has been developed taking account of the known dynamics of the process and including the missing information.