Business Solution Blog

Posts Tagged ‘Methods.’

Introduction human component as an input to the statistical forecast. The development of a statistical forecasting plays a major role in the creation of a valid demand plan, and should not be routinely treated as a black box. Christopher Koch (2004, http://www. Cio. Com/archive/061504/nike. Html), warns that “Throwing a bunch of historical sales in a program and waiting for a magic number – the basic concept behind demand software planning – not necessarily the absolute truth. ” Many planners have realized that without the human component, the statistical forecasts alone do not provide adequate estimates of demand. In an article for CIO Magazine, Ben Worthen (2003, http://www. Cio. Com/archive/071503/future. Htmlkjcv) said that “The demand forecast sounds like an exact science, but if you look carefully people are half the equation in the process. ” special issue of the journal Forseight June 2005, several authors focused their views on how to integrate quantitative and qualitative forecasts. Nigel Harvey (2005, p. 18) summed it up: “As Paul Goodwin and others, I believe that expert opinion and statistical methods complement each other in the forecasting process and that the problem for planners is to decide when and how to combine achieve the best combination. “demand Many planners realize that the automatic forecast does all the work. The statistical models used in developing their forecasts working with “raw” data. Models do not know whether the numbers represent chips or USB sticks, they are not able to interpret a slope in sales and an excess production or whether a peak in demand is the result of additional advertising or sale of a random. Moreover, statistical models do not predict unexpected circumstances. Ana Ku (2002, http://www. Analyticalq. Com / energy / demand / default. Htm) rightly mentions in his note “If the inputs to the forecasting model are poor, it would be very difficult to get a good outcome no matter how good your model.. ” is important to note that the above applies if in the process is a structured process of forecasting. Sales and Marketing – Meetings Predemanda Many processes demand planning process are the following: Generate statistical forecasts.Adjust forecasts based on market knowledge.Reaching consensus and publishing the results. A better strategy would bring knowledge as an input in the development of statistical forecasts. This strategy has been very well received among many academic outcome. J. Scott Armstrong (2001, p. 736), in his book Principles of Forecasting, mentions the importance of using knowledge of sales and marketing as inputs in the development of functional planning. Principle 11. 2: Using a structured knowledge as input to quantitative models. Principle 11. 3: Use knowledge to select, weigh, and modify the quantitative methods. investment of time and energy in models that incorporate the proper knowledge will help analysts to create more accurate statistical forecasts will reduce the number of manual changes and improves the accuracy of estimates. Company J. R. Simplot, Sales and Marketing staff participate in meetings “predemanda.” These meetings are organized, evaluated and made the latest business information. It documents the process thoroughly and sharing the results with analysts before the construction of statistical models. Here are some conclusions drawn in this process of incorporating knowledge of sales and marketing that helped to improve the statistical forecast results: When forecasting statistically SKU level and then distribute the results in descending order were achieved better results than predicted desagragada form.A classification of pareto on references to concentrate the efforts on the reference type (those critical to the business) and schedule revisions with lower frequencies in type B and C.The articles whose behavior is too variable to predict by means of a statistical model, are removed gradually as the months pass, since these items usually have special features such as contracts for fixed amounts and dates, news items which have no history, which makes the performance of a statistical model is very poor.It should measure the impact of events in the behavior of the application, identify patterns in which sales can increase suddenly and investigate the causes of these behaviors in order to ponder and predict its impact to future.It is necessary to measure performance by comparing the actual value forecast vs forecasted, identify causes and take action. is very important to have a tool to model this type of situation. Review and Consensus worth noting that qualitative adjustments will continue to be for those references which have not incorporated the history of events held (promotions, advertising etc.), or those who were not under consideration at meetings of predemanda. Once generated, the statistical forecasts are reviewed with the team’s planning application to include those new products to be predicted, which should be excluded from the statistical forecast and finally those who require additional information. As a result of this review will reach a consensus on the statistical forecast. After reaching consensus on statistical forecasting formalized scheduled reviews that involve the departments of Sales, Marketing, Operations and Management. The reasons for additional changes include new developments planned, win or lose major customers contraction of production and management actions orientadasa align sales with strategic objectives of the company and the decision to promote a particular product to achieve market share projected. Conclusion Many companies are based forecasts generated from a software to start your planning cycle demand, and then make manual adjustments incorporarn knowledge of sales, marketing, and operations. To incorporate and formalize knowledge in the construction of the base forecasts can be increased forecast accuracy between 10 and 15% and reducing manual adjustments by approximately 40%. For more information about this topic, or related issues please contact us at info @ mind. com. co or visit a ”nofollow” _trackPageview(‘/outgoing/article_exit_link’);” onclick=”javascript:pageTracker. href=”http://pronosticoexperto. com” www. pronosticoexperto. com / a, a ”nofollow” _trackPageview(‘/outgoing/article_exit_link’);” onclick=”javascript:pageTracker. href=”http://planeaciondemanda. com” www. planeaciondemanda. com / a