FORECASTING:FORECASTING IN PRACTICE

By -

FORECASTING IN PRACTICE

You now have seen the major forecasting methods used in practice. We conclude with a brief look at how widely the various methods are used.

Every company needs to do at least some forecasting, but their methods often are not as sophisticated as with these major projects. Some insight into their general approach was provided by a survey conducted some years ago2 of sales forecasting practices at 500 U.S. corporations. Although this survey published in 1994 now is somewhat out of date, we believe that its results are still somewhat reflective of current forecasting practices.

This survey indicates that at that time, judgmental forecasting methods were some- what more widely used than statistical methods. The main reasons given for using judgmental methods were accuracy and difficulty in obtaining the data required for statistical methods. Comments also were made that upper management is not familiar with quantitative techniques, that judgmental methods create a sense of ownership, and that these methods add a commonsense element to the forecast.

Among the judgmental methods, the most popular was a jury of executive opinion. This was especially true for companywide or industry sales forecasts but also holds true by a small margin over manager’s opinion when forecasting sales of individual products or families of products.

Statistical forecasting methods also are fairly widely used, especially in companies with high sales. Compared to earlier surveys, familiarity with such methods is increasing. (Given that statistical forecasting methods now have been regularly taught in business schools and management seminars for many years, we anticipate that this trend of in- creasing familiarity with such methods has continued since the time of the survey.) How- ever, many survey respondents cited better data availability as the improvement they most wanted to see in their organizations. The availability of good data is crucial for the use of these methods. (Fortunately, the rapid advances in information technology since the survey was conducted has led to much better data availability in many companies.)

The survey indicates that the moving-average method and linear regression were the most widely used statistical forecasting methods. The moving-average method was more popular for short- and medium-range forecasts (less than a year), as well as for forecasting sales of individual products and families of products. Linear regression was more popular for longer-range forecasts and for forecasting either companywide or industry sales. Both exponential smoothing and the last-value method also received considerable use. However, the highest dissatisfaction is with the last-value method, and its popularity was decreasing compared to earlier surveys.

When statistical forecasting methods were used, it was fairly common to also use judgmental methods to adjust the forecasts. (This continues to be fairly common practice.)

As managers have continued to become more familiar with statistical methods, and more used to using the computer to compile data and implement OR techniques, we anticipate that the usage of statistical forecasting methods is continuing to grow. However, there always will be an important role for judgmental methods, both alone and in combination with statistical methods.

Another important trend in recent years has been an increasing availability and usage of sophisticated software packages for applying statistical forecasting methods. (See Selected Reference 11 for a survey of these packages.) Selected Reference 10 also provides a survey of the use, satisfaction, and performance of forecasting software in practice. Most of the U.S. corporations responding to the latter survey reported using software for their forecasts, although this sometimes involved using only spreadsheets or internally developed forecast- ing software. Those using commercially available software packages reported both the best forecasting performance and the greatest satisfaction with the features of the software.

Comments

Post a Comment

Popular posts from this blog

DUALITY THEORY:THE ESSENCE OF DUALITY THEORY

By -
Image
THE ESSENCE OF DUALITY THEORY Given our standard form for the primal problem at the left (perhaps after conversion from another form), its dual problem has the form shown to the right. Thus, with the primal problem in maximizatio n form, the dual problem is in minimization form instead. Furthermore, the dual problem uses exactly the same pa r amete r s as the primal problem, but in different locations, as summarized below. 1. The coefficients in the objective function of the primal problem are the right-hand sides of the functional constraints in the dual problem. 2. The right-hand sides of the functional constraints in the primal problem are the coef- ficients in the objective function of the dual problem. 3. The coefficients of a variable in the functional constraints of the primal problem are the coefficients in a functional constraint of the dual problem. To highlight the comparison, now look at these same two problems in matrix notation (as introduced at the beginn...
Read more

NETWORK OPTIMIZATION MODELS:THE MINIMUM SPANNING TREE PROBLEM

By -
Image
THE MINIMUM SPANNING TREE PROBLEM The minimum spanning tree problem bears some similarities to the main version of the shortest-path problem presented in the preceding section. In both cases, an undi r ecte d and connected network is being considered, where the given information includes some mea- sure of the positive length (distance, cost, time, etc.) associated with each link. Both problems also involve choosing a set of links that have the shortest total length among all sets of links that satisfy a certain property. For the shortest-path problem, this property is that the chosen links must provide a path between the origin and the destination. For the minimum spanning tree problem, the required property is that the chosen links must provide a path between ea c h pair of nodes. The minimum spanning tree problem can be summarized as follows: 1. You are given the nodes of a network but not the links. Instead, you are given the po- tential links and the positive length for each ...
Read more

INTEGER PROGRAMMING:THE BRANCH-AND-CUT APPROACH TO SOLVING BIP PROBLEMS

By -
TH E BRANCH-AND-CUT APPROACH TO SOLVING BIP PROBLEMS Integer programming has been an especially exciting area of OR since the mid-1980s be- cause of the dramatic progress being made in its solution methodology. Background To place this progress into perspective, consider the historical background. One big break- through had come in the 1960s and early 1970s with the development and refinement of the branch-and-bound approach. But then the state of the art seemed to hit a plateau. Rel- atively small problems (well under 100 variables) could be solved very efficiently, but even a modest increase in problem size might cause an explosion in computation time be- yond feasible limits. Little progress was being made in overcoming this exponential growth in computation time as the problem size was increased. Many important problems arising in practice could not be solved. Then came the next breakthrough in the mid-1980s, with the introduction of the b r an c h- and-cu t approach to sol...
Read more