Spare part demand forecasting for consumer goods using installed base information

Abstract

Full bibliographic details must be given when referring to, or quoting from full items including the author's name, the title of the work, publication details where relevant (place, publisher, date), pagination, and for theses or dissertations the awarding institution, the degree type awarded, and the date of the award.

Forecasting spare part demand with installed base information: A review

“…In the mature phase, where product sales gradually fall back and the installed base is at its maximum size, the spare part demand is expected to rise. In the end-of-life (EOL) phase, where product sales end, the demand for spare parts can still continue to increase, before it gradually diminishes as more products reach their end-of-use (Kim et al, 2017). From this figure it can be seen that the demand for spare parts follows the demand for the product, albeit with a delay.…”

“…Hong et al (2008) add the replacement probability of the failed part, because a failed part is not necessarily replaced, or it might be replaced by an unauthorized part as well. Dekker et al (2013) and Kim et al (2017) argue that the demand for spare parts follows the demand for the installed product with a delay. It is thus dependent on the product life-cycle.…”

“…The installed base of a product is the number of sold products that can lead to demand of their spare parts (Kim et al, 2017). Dekker et al (2013) stress the importance of knowing the characteristics of this installed base (e.g., age, usage) to make correct inventory decisions.…”

Automobile spare-parts forecasting: A comparative study of time series methods

“…Rosienkiewicz et al [43] obtained better results by using ANNs than with any other time series method, just as we did for the sample periods. Dekker et al [17] and Kim et al [18], found good forecasts using the installed base information, but gave no comparison against the time series models. It would be strongly advisable to distinguish the results for the sample and postsample periods in any similar forecasting research study.…”

“…This approach has also been utilised for consumer goods, suggesting that this is a very good way to forecast the demand for spare parts, especially at the end of the life of the product, but also recommends to use simpler methods if the proper information is not currently available [18]. Parallel to this technique that uses no time series, a forecasting method in two steps has been proposed, which separately updates the average of the parts needed per repair and the number of repairs for each sort of component.…”