Advanced computational methodologies on open big data for forest terrain trafficability monitoring and forecasting

The proposed project is focused to the development of big open data and computational methodologies to monitor and estimate forest
terrain trafficability which are needed to a) downscale regional predictions to smaller scale, b) improve and develop methods for big
data – trafficability analysis --> prediction models, c) create cloud computing based computational and data storage environment
making possible to collect in situ data from operational forest machines. Such a framework would allow collection of data from range of
sites, different machines etc., and constant development (and re-calibration) of the prediction models. This is done by utilizing different
types of advanced observational open data sources (including satellite images and weather information) and probabilistic and
information theoretical data analysis and assimilation approaches employing multi-source heterogeneous spatiotemporal information.
An important contribution of the project is establishing a cloud computing environment allowing feedback to/from in situ observations
by forest harvesters. With increasing number of open data sources combined with relevant cloud computing services, big data can
become a major factor of competitiveness of forest based bioeconomy.

Projektin valmistelu on käynnistetty tutkimus- ja asiakkuusjohtaja Jari Varjon aloitteesta