Since the Linear Programming(LP) was introduced to forest management planning, many timber harvest scheduling models have been developed, such as Timber RAM(Resource Allocation Method : Navon, 1971), MUSYC(Multiple Use Sustained Yield Calculation: Johnson, 1979), and FORPLAN(FORest PLANning : Johnson, 1980), etc. Because these model; were designed to develop timber harvest schedules that would be biologically sustainable while maximizing an objective of timber volume or the present net worth, they differ from and have more complex calculation elements than the previous LP models that maximize only timber volume. Among them, Timber RAM has been called Model I by Johnson and Scheurman(1977) and has been widely used to address issues related to biological sustain ability of timber harvests. This study was conducted to investigate Timber RAM calculation elements in detail, and to present harvest and regulation method by the model, and to apply Timber RAM to the practical situation in Kangweon Natffll. Univ. Exp. Forests. Timber RAM is a computerized method for generating long range forest planning under multiple-use management and can span up to 35 decades. A distinctive feature of this model is to determine an optimum cutting schedule while controlling the area or volume harvested for each decade during the CP (Conversion Period) and PCP (Post Conversion Period). Within the model, decisions are made at the timber class level, with constraints reflecting overall forestwide goals and objectives. The areas of existing timber regeneration harvested for each period in a timber class form a management unit whose integrity is retained throughout the PH (Planning Horizon). Seventeenth runs under different constraints were tried 2o examine the applicability of Timber RAM, using the volume control and regulation option in harvest constraints, and the sequential lower and upper bounds in harvest control constraints. The results of application indicated that if the wider control and regulation range of periodic harvest volume during the CP and PCP were adopted, total harvest volume increased but the area and volume harvested for each period showed greatly fluctuation, whereas if the narrower regulation range of periodic harvest volume during the PCP was adopted, total harvest volume decreased but the area and volume harvested for each period showed comparatively small fluctuation. And in Timber RAM, there was no flexibility for harvesting regenerated timber. After the existing timber is harvested, the area must be reharvested at the rotation age of the regenerated timber. It is required that a large variation in the PCP should be allowed because of the lack of flexibility in Timber RAM for handling regenerated timber. The decision to use 99% in the PCP for the base runs should be viewed as a solution to the minor problem with Timber RAM rather than as a harvest scheme being seriously proposed for the PCP. LRSYA (Long Range Sustained Yield Average) is equivalent to the long range productivity of the forest. It is a constant for given timber class areas, regenerated yield, and rotations and is determined by calculating the area-weighted mean annual increment of the regenerated timber class yields. It plays an essential role in allowable cut determination if sustained yield is important. That is, if the RAM-determined allowable cut is greater than LRSYA, the allowable cut must decrease at some time in the future, whereas if the cut is less than LRS YA, the allowable cut must increase over time, The study indicate that, since the CP and most PCP average harvests are less than LRSYA, the allowable cut must increase at some time in the future. Although it is desirable that most of the harvests of existing timber occur in the CP while most of the harvests in the PCP come from regenerated timber, the CP and PCP seem to be arbitrary divisions in Timber RAM. But these divisions for the PH greatly simplify the structure of the computer program, In