An aggregate–disaggregate intermittent demand approach (ADIDA) to forecasting: an empirical proposition and analysis

Abstract

Intermittent demand patterns are characterised by infrequent demand arrivals coupled with variable demand sizes. Such patterns prevail in many industrial applications, including IT, automotive, aerospace and military. An intuitively appealing strategy to deal with such patterns from a forecasting perspective is to aggregate demand in lower-frequency 'time buckets' thereby reducing the presence of zero observations. However, such aggregation may result in losing useful information, as the frequency of observations is reduced. In this paper, we explore the effects of aggregation by investigating 5,000 Stock Keeping Units (SKUs) from the Royal Air Force (RAF, UK). We are also concerned with the empirical determination of an optimum aggregation level as well as the effects of aggregating demand in time buckets that equal the lead time length (plus review period). This part of the analysis is of direct relevance to a (periodic) inventory management setting where such cumulative lead-time demand estimates are required. Our study allows insights to be gained into the value of aggregation in an intermittent demand context. The paper concludes with an agenda for further research in this area.

Forecasting with temporal hierarchies

“…Rostami-Tabar et al (2013) look into identifying an optimal aggregation level when the disaggregate process is either MA(1) or AR(1) and find that in general the higher the aggregation the lower the forecast errors are, when forecasting by single exponential smoothing. For more complex processes Nikolopoulos et al (2011) demonstrate that forecasting accuracy does not change monotonically as the aggregation level increases.…”

Supply chain forecasting: Theory, practice, their gap and the future

“…Temporal aggregation is known to be applied widely in practical military settings (very sparse data), the after-sales industry (service parts) and elsewhere. Nikolopoulos et al (2011) demonstrated the value of (non-overlapping) temporal aggregation in such a context. Although the paper under concern was exploratory in nature, it provided some much-needed empirical evidence of the potential benefits of using temporal aggregation in a supply chain context by means of experimenting with the demand histories of 5,000 stock keeping units from the Royal Air Force (UK).…”

“…Building on those empirical results, some theoretical analysis was presented for data following an AR(1) and MA(1) process by Rostami-Tabar et al (2013) and a more general ARMA(1,1) process by Rostami-Tabar et al (2014). This analysis led to theoretically determined optimum aggregation levels for the ADIDA framework (as opposed to the empirically determined optimum levels of aggregation suggested by Nikolopoulos et al, 2011) under specific demand generation processes). Further theoretical developments for both fast and slow moving data were presented by Spithourakis et al (2014) providing theoretical bounds for the temporal aggregation process to be leading to improvements in forecasting accuracy.…”

“…This is very important if we take into account that in many application areas we are interested in a cumulative forecast. This issue has been considered by Nikolopoulos et al (2011) providing evidence that tying the forecasting horizon to the lead time could lead to considerable forecast accuracy improvement, when measuring performance at that level of aggregation (i.e. equal to the lead time) (see also Nikolopoulos and Petropoulos, 2015, for a discussion of this issue in the context of high uncertainty).…”

Previsión de demanda intermitente con métodos de series de tiempo y redes neuronales artificiales: Estudio de caso

“…Conocido como método ADIDA (Aggregate, Disaggregate, Intermittent Demand Approch, por sus iniciales en inglés) transforma la serie inicial en otra serie modificando la frecuencia en que se agregan las demandas de esos periodos, por ejemplo, de mensual a trimestral. Con la serie alternativa se realiza la extrapolación y luego se procede con la desagregación para estimar la previsión de la demanda para un determinado período [15].…”