Demand forecasting by temporal aggregation

Demand forecasting performance is subject to the uncertainty underlying the time series an organization is dealing with. There are many approaches that may be used to reduce uncertainty and thus to improve forecasting performance. One intuitively appealing such approach is to aggregate demand in lower‐frequency “time buckets.” The approach under concern is termed to as temporal aggregation, and in this article, we investigate its impact on forecasting performance. We assume that the nonaggregated demand follows either a moving average process of order one or a first‐order autoregressive process and a single exponential smoothing (SES) procedure is used to forecast demand. These demand processes are often encountered in practice and SES is one of the standard estimators used in industry. Theoretical mean‐squared error expressions are derived for the aggregated and nonaggregated demand to contrast the relevant forecasting performances. The theoretical analysis is supported by an extensive numerical investigation and experimentation with an empirical dataset. The results indicate that performance improvements achieved through the aggregation approach are a function of the aggregation level, the smoothing constant, and the process parameters. Valuable insights are offered to practitioners and the article closes with an agenda for further research in this area. © 2013 Wiley Periodicals, Inc. Naval Research Logistics 60: 479–498, 2013     

Worst‐case performance of approximation algorithms for tool management problems

Since the introduction of flexible manufacturing systems, researchers have investigated various planning and scheduling problems faced by the users of such systems. Several of these problems are not encountered in more classical production settings, and so‐called tool management problems appear to be among the more fundamental ones of these problems. Most tool management problems are hard to solve, so that numerous approximate solution techniques have been proposed to tackle them. In this paper, we investigate the quality of such algorithms by means of worst‐case analysis. We consider several polynomial‐time approximation algorithms described in the literature, and we show that all these algorithms exhibit rather poor worst‐case behavior. We also study the complexity of solving tool management problems approximately. In this respect, we investigate the interrelationships among tool management problems, as well as their relationships with other well‐known combinatorial problems such as the maximum clique problem or the set covering problem, and we prove several negative results on the approximability of various tool management problems. © 1999 John Wiley & Sons, Inc. Naval Research Logistics 46: 445–462, 1999     

Partially controlled demand and inventory control: An additive model

The primary goal of this paper is to establish properties of the inventory and advertising policy minimizing the expected discounted cost over a finite horizon in a dynamic nonstationary inventory model with random demand which is influenced by the level of goodwill. Under linearization of the cost associated with the maximum inventory and the advertising effect on demand, the model is shown to be equivalent to an inventory model with disposal. Many results of this paper are extended to cover convex ordering cost of inventory and time lag in delivery of stocks.     

A note on the forecast performance of temporal aggregation

Earlier research on the effects of nonoverlapping temporal aggregation on demand forecasting showed the benefits associated with such an approach under a stationary AR(1) or MA(1) processes for decision making conducted at the disaggregate level. The first objective of this note is to extend those important results by considering a more general underlying demand process. The second objective is to assess the conditions under which aggregation may be a preferable approach for improving decision making at the aggregate level as well. We confirm the validity of previous results under more general conditions, and we show the increased benefit resulting from forecasting by temporal aggregation at lower frequency time units. © 2014 Wiley Periodicals, Inc. Naval Research Logistics 61: 489–500, 2014     

Optimal advertising and inventory control of perishable goods

The primary goal of this article is to extend the results of a previous article to the case where the effect of advertisement on sales lasts more than one period. Monotonicity of the optimal advertising and inventory policies in the various factors is investigated. Also, attention will be focused on the relationship between the fluctuations over time of the optimal policies and the variations over time of the factors involved, such as demand distributions and holding costs. For example, if over a finite interval the demand decreases from one period to the next, reaches a minimum, and then increases, then the optimal advertising policy will produce the opposite effect. The period of minimum demand never precedes that of maximum goodwill; moreover, the optimal inventory level decreases while the demand decreases. Finally, when demand distributions are just translates of one another, the results of this article can be extended to nonperishable goods.     

Fluctuations of the optimal advertising and inventory policy: A qualitative analysis

This paper discusses the properties of the inventory and advertising policy minimizing the expected discounted cost over a finite horizon in a dynamic nonstationary inventory model with random demand which is influenced by the level of promotion or goodwill. Attention is focused on the relation between the fluctuations over time of the optimal policies and the variations over time of the factors involved, i.e., demand distributions and various costs. The optimal policies are proved to be monotone in the various factors. Also, three types of fluctuations over time of the optimal policies are discussed according to which factor varies over time. For example, if over a finite interval, the random demand increases (stochastically) from one period to the next, reaches a maximum and then decreases, then the optimal inventory level will do the same. Also the period of maximum of demand never precedes that of maximum inventory. The optimal advertising behaves in the opposite way and its minimum will occur at the same time as the maximum of the inventory. The model has a linear inventory ordering cost and instantaneous delivery of stocks; many results, however, are extended to models with a convex ordering cost or a delivery time lag.