The purpose of this thesis was to develop the concept of cost-efficient Juvenile Stand Management (JSM) for planted Norway spruce (Picea abies L. Karst) stands. The principles of time based management were followed, by integrating regeneration activities as a cost-efficient value chain and by minimizing non-value-adding work with straightforward decision making based on forest management plan data.The effects of soil preparation and Early Cleaning (EC) on further development of the stands were studied in intensive field experiments. Extensive survey data were used to develop methods applicable for efficient decision making in JSM, such as estimating need for EC or labor time consumption of PreCommercial Thinning (PCT).Timing of JSM had major effect on its costs; a delay in PCT increased the labor time needed to manage a stand by 8.3% annually. Moreover, 61-70% of the saplings in a typical Norway spruce stand were considered to need EC years before PCT was appropriate to be done. EC was also found to be an effective release treatment as it subsequently increased the diameter growth of crop trees by 21-32%. However, a two-stage management regimen, which included EC and PCT, appeared to be somewhat more labor consuming than the PCT only option. Soil preparation method had a major effect on emergence and growth of non-crop trees, and thus, on overall costs of JSM-program. The results showed that understanding the interactions in regeneration chain activities is important for productive forestry.Furthermore, a priori information can have practical implications in decision making for JSM. Several site or stand attributes were found to explain labor consumption of PCT or the need for EC. However, decision making in daily forestry requires more reliable models. The modelling data should go beyond the data of traditional forest management planning in further research. Big data offers promising opportunities. Foremost, I'm grateful to Timo Saksa who supported me in every situation necessary during this work and encouraged me to finally gain this achievement. I'm very grateful to Juho Rantala for his help, guidance and encouragement, which was invaluable for me as a researcher and for this thesis. Without these two people this work would not have been happened. I'm thankful also to my other supervisors, valuable comments from Pasi Puttonen and Lauri Valsta gave much content for the thesis and made it more coherent. I like to express my gratitude to Juha Lappi, a good colleague next to my room who has been eager to help me with statistics in any moment I've needed it. I'm grateful to my co-authors Pertti Harstela and Nuutti Kiljunen for their help in planning the research and commenting the articles. I'm thankful to Pekka Rossi, Pekka Voipio, Mervi Seppänen, Jussi Nuutinen, and Mikko Nykänen for assisting with field experiments, and to Alisdair McLean and Michael Hardman for revising the English language. I would also like to express my thanks to pre-examiners Doug Pitt and Kari Mielikäinen for their valuable comments.I'm g...