Vassilios Assimakopoulos

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

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 5000 stock keeping units from the Royal Air Force (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.

‘Horses for Courses’ in demand forecasting

Forecasting as a scientific discipline has progressed a lot in the last 40years, with Nobel prizes being awarded for seminal work in the field, most notably to Engle, Granger and Kahneman. Despite these advances, even today we are unable to answer a very simple question, the one that is always the first tabled during discussions with practitioners: “what is the best method for my data?”. In essence, as there are horses for courses, there must also be forecasting methods that are more tailored to some types of data, and, therefore, enable practitioners to make informed method selection when facing new data. The current study attempts to shed light on this direction via identifying the main determinants of forecasting accuracy, through simulations and empirical investigations involving 14 popular forecasting methods (and combinations of them), seven time series features (seasonality, trend, cycle, randomness, number of observations, inter-demand interval and coefficient of variation) and one strategic decision (the forecasting horizon). Our main findings dictate that forecasting accuracy is influenced as follows: (a) for fast-moving data, cycle and randomness have the biggest (negative) effect and the longer the forecasting horizon, the more accuracy decreases; (b) for intermittent data, inter-demand interval has bigger (negative) impact than the coefficient of variation; and (c) for all types of data, increasing the length of a series has a small positive effect.

Empirical heuristics for improving intermittent demand forecasting

Purpose – Intermittent demand appears sporadically, with some time periods not even displaying any demand at all. Even so, such patterns constitute considerable proportions of the total stock in many industrial settings. Forecasting intermittent demand is a rather difficult task but of critical importance for corresponding cost savings. The current study aims to examine the empirical outcomes of three heuristics towards the modification of established intermittent demand forecasting approaches.Design/methodology/approach – First, optimization of the smoothing parameter used in Crostons approach is empirically explored, in contrast to the use of an a priori fixed value as in earlier studies. Furthermore, the effect of integer rounding of the resulting forecasts is considered. Lastly, the authors evaluate the performance of Theta model as an alternative of SES estimator for extrapolating demand sizes and/or intervals. The proposed heuristics are implemented into the forecasting support system.Findings – Th...

Architecture for a real estate analysis information system using GIS techniques integrated with fuzzy theory

Real Estate analysis in urban area is an important and very difficult task. The value of a property is primarily determined by its location. Thus a geographical information system (GIS) is of great importance in the evaluation process. In the last decade a very promising approach of dealing uncertainty in real estate analysis is Fuzzy theory due to the need of handling the large number of qualitative and quantitative variables that affect the value of an estate. This paper discusses the architecture for a decision support system for Real Estate analysis based on GIS techniques integrated with Fuzzy theory. In order to evaluate the system data were used from the urban area of P. Faliro in Attica basin of Greece.

An expert system for forecasting mutual funds in Greece

Forecasting the returns from investments in mutual funds is a very difficult problem. This study examines a new forecasting approach and system for the performance of mutual funds in Greece. This is accomplished via an application of a variation of the Theta model on a time series composed of the daily values of mutual funds. The proposed models are simple and implemented into an easy-to-use expert forecasting system.

Improving the performance of popular supply chain forecasting techniques

This article empirically investigates the extension of the use of an aggregation-disaggregation forecasting approach for intermittent demand (ADIDA) to fast-moving demand data, addressing the need of supply chain managers for accurate forecasts. After a brief introduction to the framework and its background, an experiment is set up to examine its performance on data from the M3-Competition. The relevant forecasting methodology and in-sample optimization techniques are described in detail, as well as the core experimental structure and real data. Empirical results of forecasting accuracy performance are presented and discussed, placing further emphasis on the managerial implications of the framework’s being a simple, cost-efficient, and universally implementable forecasting method self-improving mechanism. Finally, all conclusions are summarized and guidelines for prospective research are proposed.

The Theta model:an essential forecasting tool for supply chain planning

The Theta model created a lot of interest in academic circles due to its surprising performance in the M3-competition, the biggest ever time series forecasting competition. As a result in the subsequent years it became a benchmark in any empirical forecasting exercise and an essential tool for efficient Supply Chain Management ad planning as it provides very accurate point forecasts. The present study focuses on if the Theta model is a special case of Simple Exponential Smoothing with drift (SES-d). The Theta model outperforms SES-d in the Quarterly-M3 and Other-M3 subsets by 0.30% and 0.36%.

Remote access to group expertise: an essential e-operation for SMEs

Remote access for Small and Medium Enterprises (SMEs) to group expertise is a key service for efficient decision support. E-tools that could provide such services would enhance the decision-making process and at the same time provide a value-for-money solution; also they are very easy to assemble. E-tools reserve very limited time of their daily workload, thus facilitating participation in such decision support tasks; not to mention that virtual attendance reduces the impact of group bias. This study presents a web-based application of a semi-Delphi approach, using modern three-tier architecture, providing real-time group decision making features.

Two-dimensional colour pattern load analysis: a tool supporting demand-side management

Abstract Load analysis is one of the most important operations that support demand-side management in large electric utilities. Ordinary load analysis techniques stress on statistical processing of hourly load data along predefined time axes, producing numerical results of a standard granularity, such as daily or weekly mean loads. In order to overcome these limitations, a new approach or load analysis was developed, based on applying a two-dimensional formulation of the hourly load data. The tables holding energy consumption values, where columns represent the days of the selected period and lines represent the 24 hours of the day, are then illustrated through the use of colour patterns. In such a way, chronological typical units of variable structure and granularity can be identified and provide the basis for an extensive cross-examination, resulting in optimized decision making and energy policy definition. In order to demonstrate the advantages of the approach, a dedicated DSS implementation and application in the Greek public power corporation was also performed.

Statistical and Machine Learning forecasting methods: Concerns and ways forward

Machine Learning (ML) methods have been proposed in the academic literature as alternatives to statistical ones for time series forecasting. Yet, scant evidence is available about their relative performance in terms of accuracy and computational requirements. The purpose of this paper is to evaluate such performance across multiple forecasting horizons using a large subset of 1045 monthly time series used in the M3 Competition. After comparing the post-sample accuracy of popular ML methods with that of eight traditional statistical ones, we found that the former are dominated across both accuracy measures used and for all forecasting horizons examined. Moreover, we observed that their computational requirements are considerably greater than those of statistical methods. The paper discusses the results, explains why the accuracy of ML models is below that of statistical ones and proposes some possible ways forward. The empirical results found in our research stress the need for objective and unbiased ways to test the performance of forecasting methods that can be achieved through sizable and open competitions allowing meaningful comparisons and definite conclusions.