期刊名称：Vojnotehnicki glasnik / Military Technical Courier
印刷版ISSN：0042-8469
电子版ISSN：2217-4753
出版年度：2012
卷号：60
期号：2
页码：235-244
语种：
出版社：Ministry of defence of the Republic of Serbia: University of defence in Belgrade
摘要：Knowing the demand of spare parts is the basis for successful spare parts inventory management. Inventory management has two aspects. The first one is operational management: acting according to certain models and making decisions in specific situations which could not have been foreseen or have not been encompassed by models. The second aspect is optimization of the model parameters by means of inventory management. Supply items demand (asset demand) is the expression of customers' needs in units in the desired time and it is one of the most important parameters in the inventory management. The basic task of the supply system is demand fulfillment. In practice, demand is expressed through requisition or request. Given the conditions in which inventory management is considered, demand can be: - deterministic or stochastic, - stationary or nonstationary, - continuous or discrete, - satisfied or unsatisfied. The application of the maintenance concept is determined by the technological level of development of the assets being maintained. For example, it is hard to imagine that the concept of self-maintenance can be applied to assets developed and put into use 50 or 60 years ago. Even less complex concepts cannot be applied to those vehicles that only have indicators of engine temperature - those that react only when the engine is overheated. This means that the maintenance concepts that can be applied are the traditional preventive maintenance and the corrective maintenance. In order to be applied in a real system, modeling and simulation methods require a completely regulated system and that is not the case with this spare parts supply system. Therefore, this method, which also enables the model development, cannot be applied. Deterministic models of forecasting are almost exclusively related to the concept of preventive maintenance. Maintenance procedures are planned in advance, in accordance with exploitation and time resources. Since the timing of preventive maintenance, the number of assets on which the preventive maintenance procedures are performed and the technology of maintaining procedures are known, then the range and the quantity of spare parts needed to perform these procedures are also known. Heuristic forecasting is related to the experts. Armed forces use it for the assessment of combat situation, taking into account the opponent's and its own action tactics, forecasting enemy's intentions, analyzing the plan of operations, making decision for the plan of action, etc. In this case, relatively unregulated systems are those in which there is no data about the observed phenomenon and its development in the past. In addition to the fact that the above data does not exist, there is a need for spare parts demand forecasts for the purpose of decision making and inventory management. This primarily relates to the period of transition from an unregulated to a regulated system. In this case, forecasting has a limited range. No forecasting can be done for a longer period; the forecasting is reduced to the next relevant interval, i.e. to the interval relevant for the system. The application of another, relatively simple model which uses computer techniques has no particular limitations, but for that reason its results are time-limited; results are obtained only for the subsequent relevant period. The results of this model have a very limited range in planning. This model is applicable mainly in unregulated systems. It is suitable for so-called condition 'clearing'. After one or two closed cycles, a situation is brought to order, but then much more sensitive models are needed.
