EN 62550-2017 pdf download.Spare parts provisioning ( IEC 62550 : 2017 ).
The demand is expressed by the required quantity of an item during a specified period of time.
If the level of the future demand is known and can be determined, it is regarded as deterministic, otherwise it is stochastic. A demand or the distribution of demand, which is variable over time, is regarded as a dynamic process. When there is no influence of time on the demand, the process is regarded as stationary. A relatively even structure of demand will result in a continuous spare parts requirement: fluctuating demand Is regarded as sporadic. For spare parts, all described demand characteristics are applicable. A generally preferred method of demand forecast is not available.
For technical systems or facilities, a demand occurs when an item fails and has to be replaced by a functional item. The demand is determined by the intensity of use, the failure behaviour of the item and environmental aspects. Ideally, quantified values for applicable parameters are able to be determined within this context. Determination of the failure rate is achieved via investigation of the behaviour of a representative number of identical items or items of similar design by applying statistical methods, for example, Weibull analysis (IEC 61649). The prognosis method actually used for demand forecast will depend on the actual information available. If the failure rate and/or intensity of use are missing as concrete figures. only prognosis methods can be applied by substituting the missing information using assumptions with an associated higher risk.
5.2 Forecast based on consumption data
5.2.1 General
A production process can generally be regarded as a deterministic process since material planning is conducted by the production process and interrelated inventory. Buffer stocks have to be put in place only for short-term and medium-term deviations in the production. Major changes in the production demand will cause a change in production capacity. The requirements at spares stocks are mostly of a stochastic nature, i.e. only over a longer period or over a larger number of operational systems can the need be planned on the basis of average values, The deviations from these average values are often considerable.
The intermittent nature of spare part demand due to failures makes it difficult to apply conventional time series based algorithms. The choice of the right model becomes essential arid that would depend on usage rates and operational patterns.
5.2.2 Procedures for forecast
The overview of existing forecast procedures described in detail in Annex A cannot only be applied for the forecast of spare parts but also In general for the demand forecast In industry and trade.
• stress intensity;
• environmental factors.
For more detailed information, there are specific standards or reliability data handbooks for specific industries available (see IEC 61709).
Consideration could also be given to:
• supplier requirements on maintenance methods and frequencies, as well as supplier exchange frequencies for wear-out parts related to supplier guarantees;
• specific insurance requirements;
• spare parts for systems/components with long life, long usage period when technical development, or other factors (e.g. supplier out of business) affect the availability of spare parts;
• aging of spare parts/consumables in storage and requirements of correct storage — packaging, environmental characteristics such as humidity and temperature (see
IEC PAS 62435).
5.3.2 Prediction of failure rates and failure intensities
The failure rate and intensity of an item are essential parameters when determining the number of spares required. According to their definitions, failure rate is for non-repairable items and failure Intensity is used for repairable items. For this document, since it is not always possible to determine which one is applicable, failure rate is used for both repairable and non-repairable items because in both cases an action needs to be taken to deal with the failure, i.e. either a permanent replacement or a repair, The accuracy of the failure rate value is a risk factor that should be considered when estimating the quantity of spares The failure rate can be obtained from published data, historical data, testing, and operational usage. Commercial organizations publish ‘average’ values for component failure rates and environmental factors based upon data received from a wide range of industries. Historical data by a company for an item (or similar items) can produce more accurate failure rates if there is sufficient data available for the items operating in the specified or comparable environment. Testing a system to establish accurate failure rates can be prohibitively expensive and usually only serves to give confidence in the derived values. Operational usage can provide the best data, but this is not available for Initial spares. However, gathering this data can enable the failure rates to be adjusted for future spares provisioning (see also
IEC 61709).
When calculating the number of spares, it is usually assumed that the failure rate does not change over time (or number of cycles, distance travelled, etc.). This constant failure rate may not be correct, but when a large number of components is used in the product (system), each having failures that occur stochastically over time, the product (system) tends to a constant failure rate. EN 62550-2017 pdf download.