Forecasting in Daimler Chrysler - Essay Example

Forecasting in Daimler Chrysler A company of the size and sophistication of Daimler Chrysler has to use a judicious mix of judgmental and statistical forecasting methods for various aspects of its business (Armstrong, 2005). Mature business operations, which have stable and known relationships, with valid historical data for independent variables, lend themselves to quantitative forecasting methods. New areas of business need stochastic forecasting, and there are large elements of technical and specialized judgment that Daimler Chrysler must deploy for such matters. Daimler Chrysler has a healthy mix of product lines, including brands that are icons and very creative ideas which are on the brink on becoming commercial successes. Forecasting is therefore a complex business in the company, which effectively has more than one personality. Seasonal Forecasting Seasonal forecasting is less significant and used at Daimler Chrysler than for example in the energy industry, in which the seasons make for dramatic and cyclical shifts in demand. It is not as though automobile sales do not peak at certain times of the year, but our regression models are able to take care of these models. We can also use moving annual totals for monthly future projections, rather than a calendar method, to ensure that we account for relatively minor seasonal fluctuations in demand. Seasonal forecasting has some applications for our finance business. Customers of the Daimler Chrysler Bank may have seasonal fluctuations in demand for their products and services. The financial services wing of the company has to evaluate the validity of revenue forecasts in applications for loans in such cases. There are also some key materials that we use in production, which experience seasonal fluctuations in pricing and availability. Energy is an example for plants in the northern hemisphere. Leather, though relatively small in total cost, is another example, the purchase and production support services functions of the company have to keep seasonal forecasting methods in mind, when planning for supplies at optimal prices and in time. Delphi Delphi is a key component of forecasting methodology in the automobile business. Product demand is significantly affected by cultural and geo-political matters, which lie outside the domain of expertise of a car and truck maker. The company has long gestation periods for developing new products. Hence, there is an important and frequent need to use third-party consultant from diverse fields to understand future automobile needs in various parts of the world. Daimler Chrysler has also to use Delphi as it spreads its wings to new territories. Russia, China and India are amongst the new markets with high growth potential for automobile sales, but which have driving, traffic and customer needs very different from the traditional strong holds of Western Europe and the United States. Relevant trend data are lacking for these key opportunities, and Delphi provides important learning inputs for entry and consolidation strategies for such markets. The Delphi technique has a number of drawbacks (Evans, 2002, p 366). The kind of experts selected for the study affect the outcome, and there is inevitable subjectivity in their forecasts. Most of them are highly biased and may take extreme positions to suit their private agendas. The questionnaires we deploy during the interviews are a bit vague, and the responses are therefore not universally comparable. There is no valid measure of error as with regression models. Nevertheless, this method is the best when we need ideas about the distant future and how customer choices and competitive technologies are likely to change. Daimler Chrysler cannot do without Delphi, its weaknesses notwithstanding, because innovation lies at the heart of our competitive position (Daimler-Chrysler Creates New Management Model, 2006). We need expert inputs on matters such as future consumer needs and energy security. The jatropha project in India and the driver assist system are examples of new developments which we could not have initiated without Delphi exercises. There are also the new territorial reasons mentioned earlier. Technological Forecasting Our Delphi processes cover technological forecasting, with one difference: we tend to use third party experts for Delphi, whereas our own development engineers and functional specialists are involved in technological forecasting. Our Finance, Information Technology. Research and Human Resources functions participate in technological forecasting, for matters related to their domains. We also use inputs from competitors and industry associations as gleaned from seminars and similar events. Technological forecasting involves a high degree of judgment, so our senior management generally has the final say! Technological forecasting gets our personnel in to extreme positions, because each person develops strong opinions based on their individual experiences. The future of bio-fuels and their emission properties is an example of the kinds of controversies that technological forecasting can provoke. Overall, technological forecasting can provide some leads for research, but it is risky to make significant and strategic investments based on such forecasts alone. Time Series Data Daimler Chrysler revels in precision and quality engineering, and takes pride in the reliability of its products. Time series data is used for forecasting extensively, and as far as possible. Sophisticated smoothing techniques are used where the scatter of regression data is high, and the company makes a big effort to collate and store present data from dealers and other sources for future use. The new management model calls for integration, a leaner structure and focus on core business processes (Daimler-Chrysler Creates New Management Model, 2006). Mature product lines, such as buses, vans and other commercial vehicles have fairly well established demand patterns. We can use time series data for short-term sales forecasting of such product lines. Time series data is also suitable for providing demand inputs for spares and lead times for vendors in our Enterprise Resource Planning (ERP) system. Time series data is extensively used by the Finance function to monitor market capitalization and the stock market environment. This covers the value of the company's scrip, comparisons with competitors, and investments made by Daimler Chrysler in other companies as well. Major investments in new production capacities also use time series data to calculate discounted cash flows. Conclusions Daimler Chrysler is deeply involved with the future. Managing uncertainties is a key factor for success, and the company tries to excel in forecasting in order to ensure that its heritage of excellent products and superior financial performance continues in the years to come. The company relies on time series data as much as possible because it is possible to define the ranges of accuracy and probability. Delphi techniques are used when the company has to venture outside its core competencies to deal with new challenges and opportunities. The method of forecasting used has to be tailored to the business needs of each situation. Methods other than time series and Delphi have limited application in a company such as Daimler Chrysler. References Armstrong, J. S (2005), 'Forecasting Principles', Marketing Department of The Wharton School, University of Pennsylvania, retrieved February 2006 from Evans, M, (2002), Practical Business Forecasting, Blackwell Publishing 'Daimler-Chrysler Creates New Management Model', 2006, Company Website, retrieved February 2006 from Read More