Intermittent demand forecasting: a guideline for method selection

The intermittent demand forecasting problem involves the forecasting of demand series that are characterized by the time between demands being significantly larger than the unit of time used for the forecast period. This causes the time series associated with the demand to have a large percentage of periods for which there are no demands. These types of series are often found in spare parts inventory management systems. This paper examines the intermittency of a demand series by relating the lag-1 correlation coefficient of non-zero demand, squared coefficient of variation of non-zero demand and probability of zero of the demand series to the error properties of various forecasting techniques. A classification method is presented by which a time series can be characterized in terms of key parameters related to intermittency and through this relationship the best of a set of forecasting techniques can be recommended. The method is illustrated on both real intermittent demand series and randomly generated time series in order to understand the efficacy of the procedure to improve overall forecasting effectiveness.

The Journal of Engineering Research, 2021

Spare parts are among the most important and challenging fields for the intermittent demand forecast. Improving the accuracy of the forecasted results can improve the efficiency of the inventory control system as well as the adoption of beneficial strategic decisions in the face of the uncertainty inherent in the demand. The effectiveness of hybrid forecast models in predicting the stable and accurate results has made them a useful tool in counteracting the uncertainty and complexities of the time series structure. In the present study, a hybrid model will be introduced for forecasting the intermittent demand that can impressively overcome the limitations of individual models while simultaneously using the unique advantages of these models in dealing with the complexities in the intermittent demand. The accuracy in the forecasting ability is also increased by suitable examination of the structures and patterns in the intermittent data. The research modeling was done based the Crosto...

Intermittent demand items account collectively for considerable proportions of the total stock value of any organization. Forecasting the relevant inventory requirements constitutes a very difficult task and most work in this area is based on Croston's estimator that relies upon exponentially smoothed demand sizes and inter-demand intervals. This method has been shown to be biased and a number of variants have been introduced in the literature, including the recently proposed TSB method that updates the demand probability instead of the demand interval and in doing so reacts faster to decreasing demand. The TSB has been shown theoretically to be unbiased (for all points in time), but its empirical performance has not been investigated yet and this constitutes one of the objectives of our work. More generally, we explore the empirical performance of forecasting methods used in an intermittent demand context, paying particular attention to the effects and implications of the smoothing constant values employed for updating purposes. We do so by means of experimentation on large empirical datasets from the military sector and automotive industry. The results enable insights to be gained into the sensitivity of the various methods' forecasting and stock control performance to the smoothing constant values used. The paper concludes with an agenda for further research.

The South African Journal of Industrial Engineering, 2012

It is commonly assumed that intermittent demand appears randomly, with many periods without demand; but that when it does appear, it tends to be higher than unit size. Basic and well-known forecasting techniques and stock policies perform very poorly with intermittent demand, making new approaches necessary. To select the appropriate inventory management policy, it is important to understand the demand pattern for the items, especially when demand is intermittent. The use of a forecasting method designed for an intermittent demand pattern, such as Croston's method, is required instead of a simpler and more common approach such as exponential smoothing. The starting point is to establish taxonomic rules to select efficiently the most appropriate forecasting and stock control policy to cope with thousands of items found in real environments. This paper contributes to the state of the art in: (i) categorisation of the demand pattern; (ii) methods to forecast intermittent demand; and (iii) stock control methods for items with intermittent demand patterns. The paper first presents a structured literature review to introduce managers to the theoretical research about how to deal with intermittent demand items in both forecasting and stock control methods, and then it points out some research gaps for future development for the three topics.

In making forecasting, there are many kinds of data. Stationary time series data are relatively easy to make forecasting but random data are very difficult in its execution for forecasting. Intermittent data are often seen in industries. But it is rather difficult to make forecasting in general. In recent years, the needs for intermittent demand forecasting are increasing because of the constraints of strict Supply Chain Management. How to improve the forecasting accuracy is an important issue. There are many researches made on this. But there are rooms for improvement. In this paper, a new method for cumulative forecasting method is proposed. The data is cumulated and to this cumulated time series, the following method is applied to improve the forecasting accuracy. Trend removing by the combination of linear and 2 nd order non-linear function and 3 rd order non-linear function is executed to the production data of Xray image intensifier tube device and Diagnostic X-ray image processing apparatus. The forecasting result is compared with those of the non-cumulative forecasting method. The new method shows that it is useful for the forecasting of intermittent demand data. The effectiveness of this method should be examined in various cases.

2009

Mugla Journal of Science and Technology

Intermittent demand forecasting is crucial for firms and commercial activities. Recently, many researchers have focused on forecasting methods for intermittent demand and proposed various forecasting techniques. The most prominent methods among these proposed techniques are the Croston method, which is based on exponential smoothing, and its two popular variations: SBA (Syntetos-Boylan Approximation), SBJ (Shale-Boylan-Johnston Approximation). Croston method is widely used in forecasting of intermittent demand and inventory (stock) control. Since these demands usually include zero values, using the ground breaking method developed by Croston in this data becomes inevitable. Nevertheless, there are some shortcomings to this method such as producing biased forecasts and for this reason its variations have been proposed. ATA method is a recently developed forecasting method which is an alternative to exponential smoothing. In this paper we propose a modification of ATA method that can be used for forecasting of intermittent demand. We will compare the results of the proposed approach to those of Croston and other forecasting methods used for intermittent demand forecasting.

Intermittent demand is characterized by demand data that has many time periods with zero demands. It i s hard to model intermittent demand by conventional distribut ions. In previous research, an algorithm to generat e intermittent demand was developed. The algorithm generates demand based on two stages: probabilistically generatin g whether or not a demand will occur and then generating non- zero demand if appropriate. This paper reports on e fforts to utilize the demand generation procedures as an inte rmittent demand forecasting techniques called MC-ARTA-IDF- PB based on a parametric bootstrapping approach. T he parameters are probability of non-zero demand af ter zero demand, probability of non-zero demand after non-zero demand, mean of non-zero demand, non-zero demand variance and lag 1 correlation coefficient of non-z ero demands respectively. This paper compares the e ffectiveness of MC-ARTA-IDF-PB with other relevant intermittent demand forecasting techniques and evaluates i...

Mathematical Problems in Engineering, 2010

Items with irregular and sporadic demand profiles are frequently tackled by companies, given the necessity of proposing wider and wider mix, along with characteristics of specific market fields i.e., when spare parts are manufactured and sold . Furthermore, a new company entering into the market is featured by irregular customers' orders. Hence, consistent efforts are spent with the aim of correctly forecasting and managing irregular and sporadic products demand. In this paper, the problem of correctly forecasting customers' orders is analyzed by empirically comparing existing forecasting techniques. The case of items with irregular demand profiles, coupled with seasonality and trend components, is investigated. Specifically, forecasting methods i.e., Holt-Winters approach and S ARIMA available for items with seasonality and trend components are empirically analyzed and tested in the case of data coming from the industrial field and characterized by intermittence. Hence, in the conclusions section, well-performing approaches are addressed.

International Journal of Production Research, 2014

To compare different forecasting methods on demand series we require an error measure. Many error measures have been proposed, but when demand is intermittent some become inapplicable, some give counter-intuitive results, and there is no agreement on which is best. We argue that almost all known measures rank forecasters incorrectly on intermittent demand series. We propose several new error measures with wider applicability, and correct forecaster ranking on several intermittent demand patterns. We call these "mean-based" error measures because they evaluate forecasts against the (possibly time-dependent) mean of the underlying stochastic process instead of point demands.