As a forest resource rich country (73% of total land area) and a pioneer in forest technology, Finland has a long standing tradition of utilizing forest biomass for bioenergy production. The National Renewable Action Plan 2010 strives to increase the present consumption of forest chips from 6 million m(3) to 13.5 million m(3) (97 PJ) by 2020, mainly for combined heat and power (CHP) production and separate heat production. In 2013, the forest chip consumption was 8.7 million m(3). To achieve, 2020 target of forest chip consumption, Finland would still need either 8 new CHP production plants (200 MWbio) or increase the share of wood chip consumption at the existing plants (e.g., co-firing, boiler substitution, peat replaced by wood chips). The aim of this study is to apply a Mixed Integer Linear Programming model to optimize the 8 new CHP plant locations and production configurations by minimizing the entire costs and emissions of supply chain with respect to biomass resource availability, energy demand and existing industrial competition. The model results also present the feedstock resource allocation strategy with regards on the parameter sensitivity (e.g., less forest harvesting, price increase). Logging residues (92% biomass share) and sawmill residuals (8% biomass share) would be the favorite feedstock choice under current market situation. Subsidy for young thinning wood is essential for energy wood supply in future as well.