A first approach to e-forecasting: a survey of forecasting Web services

A great challenge for today’s companies is not only how to adapt to the changing business environment but also how to gain a competitive advantage from the way in which they choose to do so. As a basis for achieving such advantages, companies have started to seek to improve the performance of various operations. Forecasting is one of them; it is important to firms because it can help ensure that effective use of resources is made. In the market there are a number of off-the-shelf system products, which provide forecasts. The new trend, of moving traditional software packages to Web services, has pushed forecasting to a new dimension, named by the authors as `̀ e-forecasting’’. In this paper, a first approach to e-forecasting is made by throwing light on several aspects and a survey is presented which aims at identifying existing Web forecasting services. Nikolopoulos, 2000; Zhao et al., 2002). More accurate forecasting on the company’s monthly sales will ensure better stock policy, more efficient warehouse management, better product distribution to the company’s branches and, finally, minimization of the company’s risk in covering the market demands. Planning orders requires precision so as to reduce deviations from the final sales numbers; accurate forecasts in ordering ensure cash management and cash flow optimization. Finally, better sales forecasting ensures better exchange policy for transactions between the company and its clients. It is clear that managers can improve resource planning by understanding the limitations of forecasts. According to Wacker and Lummus (2002), these limitations are exemplified through several strategic forecasting paradoxes that managers should recognize. The paradoxes suggested by Wacker and Lummus (2002) are the: most important managerial decisions a company can make are based on the least accurate forecasts; most useful forecast information for resource planning is the least accurate; and organizations that need the most accurate forecast have the largest forecast error. y recognizing these paradoxes managers can devote their attention to improving the use and implementation of the forecast for better resource decisions. There are a number of forecasting methods or models that are available to management and the choice of technique requires a number of considerations. If management believes that the future facing their company is predictable or fairly predictable, then statistical forecasting is a useful tool. On the other hand, if a company faces a very turbulent environment where the future is mostly unpredictable, then there is little point in attempting to utilize statistical techniques to forecast the future (Lines, 1996). 3. From forecasting to e-forecasting 3.1 The definition Forecasting is evolving day-by-day. The evolutionary nature of technological development and the need for anticipating changes in technology make traditional approaches quite often obsolete. The evolution of these systems is strongly connected to Internet technology. We define e-forecasting as `̀ the ability to make forecasts by distance using the Internet’’. E-forecasting is based on the use of the Internet as the medium that is required for the delivery of the forecasting services to its clients/users. The basic concept of e-forecasting is included in the following phrase: `̀ I want to have forecasts for my data, over the Internet, only with my browser’’. 3.2 The need for e-forecasting The key question which should be answered at this stage is why companies need this new service in the context of a general e-business environment. Forecasting is, traditionally, conducted mainly through specialized forecasting software. In most cases, this software is expensive and resource-intensive in system platforms and, hence, is generally beyond the affordability of most organizations. In addition we should not underestimate the fact that this software develops its own GUI (graphical user interface) with different algorithms and methodologies, which may require welltrained staff for operation (Schubert and Ginsburg, 2000). 3.3 Main issues E-forecasting introduces major changes to the traditional way of forecasting execution. The characteristics of the Internet today impose some limitations as well as some new ways of making electronic forecasts. A first main issue is the way that data are exchanged between the client and the servers. In traditional forecasting systems, the data for the forecasting process are stored in computers provided by the client. In e-forecasting, everything is in the company providing the service. In Figure 1 we can see a simplified overview of forecasting Web services infrastructure as it can appear in an e-business environment. An e-forecasting service gives to the user the storage space for his data. All the processes take place at the provider of the service. The user can then choose his data or upload new data. In e-forecasting, the user can select in real-time the forecasting method from a set of methods that are available to him through the service’s options. Then the system is processing the data with the user’s preferences and provides the forecasts online with graphs and other graphical representation. The user finally has the ability to make some new scenarios on his data or even save these forecasts and graphical representations for later use. [ 147 ] K. Nikolopoulos, K. Metaxiotis, V. Assimakopoulos and E. Tavanidou A first approach to e-forecasting: a survey of forecasting Web services Information Management & Computer Security 11/3 [2003] 146-152