Forecasting Case

Forecasting is used for everything from global warming to the number of times a person might cough to the total sales a company will record in the next year. Forecasting can be used in just about every facet of business, and often steer the organization down the path of success or failure. Business forecasting has evolved over the years from primitive methods to very sophisticated models. Choosing the appropriate model or method to achieve the goals of the forecast is crucial to obtaining useful and meaningful information. The need for forecasting, along with its challenges as well as the various models will be explored.

Though in infancy, forecasting has existed in one form or another since nearly the beginning of time. For example, people needed to understand climates' seasonality to plan their food resources. However, prior to the 1950's, businesses had minimal systematic business forecasting tools. In the mid 1950's, exponential smoothing techniques were developed. In the latter part of the 1950's, decomposition methods were introduced. As computer costs became less expensive in the 1960's, more statistically sophisticated methods such as econometric and multiple regression models emerged. The 1970's carried in Sybil/Runner which was one of the earliest forecasting software systems. In 1976, George Box and Gwilym Jenkins presented the Box Jenkins Methods which began to unify the various theories that had previously been developed. By the early 80's the Delphi approach and cross-impact matrices were being used. The 1980's was also a period when data was "fitted" to a model, rather than the model "fitting" to the data as was the previous approach. Many of these earlier models have been refined to utilize more applicable data, which make the forecasts more meaningful. "What if models" have enhanced the value of some forecasting models. The 1990's brought the development of Enterprise Resource Planning (ERP.)

The continued advancement of technology has been the major driver for the advances in forecasting capabilities as well as the development of new and robust models. What was once a long, manual, and tedious process has now evolved into electronically updated forecasts by the minute! This has become a requirement in today's society where intolerance is the rule and waiting is unacceptable. An abundance of software packages utilizing the various forecasting models is now available. Supporting the evolution has been executive management support along with both internal and external collaboration. Input from all is essential to ensure every potential factor and influence is considered and given appropriate value. Not only is staff and expert opinion important, but customer feedback is vital.

New techniques continue to be developed. For example, Dynamic on Demand Supply Chains (DODSC) has brought four major elements into use. (1) Robust Process Improvement Methodologies operate under the guise of being "lean," reduced variability, and identification and maximization of bottlenecks; (2) Supply Chain Best Practices expect manufacturers to monitor demand and respond quickly to changes, segregation of customer types to create cost effective supply chains, digitally develop "what if" scenarios for supply chains at multiple locations globally to produce effective supply chains, automate production schedules so they are current at all times; (3) On Demand Technology studies require "pay as you go" or pay as used situations to be implemented; (4) Change Management Methodologies document the result of changes decisions.

Another newer technique is Adaptive Collaboration. This method recognizes that some methods are accurate for short term forecasting while others are accurate for long term predictions. Once the methods have been identified, it is easier to apply appropriate weights to each method.

Total demand management is becoming the forecasting emphasis thus reducing the reliance on time series forecasting models. Total demand management includes integration of systems, leverage of external data, combining quantitative and qualitative factors, and a reliance on vendor-customer collaboration. Collaboration is critical in relationships where Vendor Managed Inventory ("VMI") programs are used. VMI programs ensure that customers will have their inventory needs met at required times. In order to facilitate this capability, customers provide the suppliers with sales and usage data on demand. Collaborative Planning, Forecasting, and Replenishing (CPFR) processes are being implemented to continually review profitability for both supplier and customer. This assists in quick and accurate decision making capabilities to generate profitable sales for manufacturers and their customers. A "knowledge expert" has advanced as the leader in forecasting. This is a person who understands the models and technology, as well as "understands the business, marketplace and techniques to generate a repeatable, operational, unbiased forecast plan." The knowledge expert not only has analytical and technological abilities, but can also "build relationships with business partners." One philosophy that has emerged is automated forecasts have become a "part of the overall corporate planning effort." No one application should stand alone. All tools need to be integrated and considered when planning and forecasting.

In is important to first understand what a forecast is, as well as how it differs from a budget. A budget is a monetary summary of expected revenues and expenses. "Forecasting is the process of making statements about events whose actual outcomes (typically) have not yet been observed." Ultimately, a budget is the compilation of many forecasts. Some of the forecasts may be dependent one another. Typically, the most important forecast of a classic budget is the sales forecast, since sales are as the basis to forecast many other lines of the budget. For example, to develop the production forecast, the sales forecast must be known. The quantity to be sold is a requirement to determine how many items need to be produced. In order to create of valid budget, the preparer must thoroughly understand the strategies and goals of the organization.

In the manufacturing sector, forecasts are necessary to plan for raw materials purchases and vendor price negotiations, labor and training, capacity and space, advertising, cash, and equipment requirements. Failure to plan properly can cause a significant negative impact to the bottom line, in addition to potential cash shortfalls. If the sales forecast is too high, too many raw materials will be purchased, excess labor and overhead will be incurred, and too much space will be required, all of which