Reinforcement learning in optimizing forest management

Show simple item record

dc.contributor.author Malo, Pekka
dc.contributor.author Tahvonen, Olli
dc.contributor.author Suominen, Antti
dc.contributor.author Back, Philipp
dc.contributor.author Viitasaari, Lauri
dc.date.accessioned 2021-10-29T06:25:02Z
dc.date.available 2021-10-29T06:25:02Z
dc.date.issued 2021-10
dc.identifier.citation Malo , P , Tahvonen , O , Suominen , A , Back , P & Viitasaari , L 2021 , ' Reinforcement learning in optimizing forest management ' , Canadian Journal of Forest Research , vol. 51 , no. 10 , pp. 1393-1409 . https://doi.org/10.1139/cjfr-2020-0447
dc.identifier.other PURE: 169878106
dc.identifier.other PURE UUID: 3d3ac9d4-44ee-4759-958f-ddb50c116da4
dc.identifier.other WOS: 000706532600001
dc.identifier.uri http://hdl.handle.net/10138/335816
dc.description.abstract We solve a stochastic high-dimensional optimal harvesting problem by using reinforcement learning algorithms developed for agents who learn an optimal policy in a sequential decision process through repeated experience. This approach produces optimal solutions without discretization of state and control variables. Our stand-level model includes mixed species, tree size structure, optimal harvest timing, choice between rotation and continuous cover forestry, stochasticity in stand growth, and stochasticity in the occurrence of natural disasters. The optimal solution or policy maps the system state to the set of actions, i.e., clear-cutting, thinning, or no harvest decisions as well as the intensity of thinning over tree species and size classes. The algorithm repeats the solutions for deterministic problems computed earlier with time-consuming methods. Optimal policy describes harvesting choices from any initial state and reveals how the initial thinning versus clear-cutting choice depends on the economic and ecological factors. Stochasticity in stand growth increases the diversity of species composition. Despite the high variability in natural regeneration, the optimal policy closely satisfies the certainty equivalence principle. The effect of natural disasters is similar to an increase in the interest rate, but in contrast to earlier results, this tends to change the management regime from rotation forestry to continuous cover management. en
dc.format.extent 17
dc.language.iso eng
dc.relation.ispartof Canadian Journal of Forest Research
dc.rights cc_by
dc.rights.uri info:eu-repo/semantics/openAccess
dc.subject reinforcement learning
dc.subject forestry
dc.subject economics
dc.subject stochasticity
dc.subject continuous cover forestry
dc.subject uneven-aged forestry
dc.subject optimal rotation
dc.subject mixed species
dc.subject optimal harvesting
dc.subject UNEVEN-AGED FORESTS
dc.subject RISK
dc.subject GROWTH
dc.subject STAND
dc.subject FIRE
dc.subject OPTIMIZATION
dc.subject ECONOMICS
dc.subject ROTATION
dc.subject 4112 Forestry
dc.title Reinforcement learning in optimizing forest management en
dc.type Article
dc.contributor.organization Economic-ecological optimization group
dc.contributor.organization Helsinki Institute of Sustainability Science (HELSUS)
dc.contributor.organization Forest Economics, Business and Society
dc.contributor.organization Environmental and Resource Economics
dc.contributor.organization Department of Forest Sciences
dc.description.reviewstatus Peer reviewed
dc.relation.doi https://doi.org/10.1139/cjfr-2020-0447
dc.relation.issn 0045-5067
dc.rights.accesslevel openAccess
dc.type.version publishedVersion

Files in this item

Total number of downloads: Loading...

Files Size Format View
forest.pdf 2.041Mb PDF View/Open

This item appears in the following Collection(s)

Show simple item record