Forecasting can be broken down into four major categories: qualitative, time series analysis, causal relationships, and simulation(Chase, Jacobs & Aquilano, 2005). Qualitative forecasting is subjective and uses estimates and opinions to predict. Simulation forecasting allows its users to modify different factors and conditions about a particular event or situation to arrive at the prediction of future conditions. Time series analysis uses historical data to predict future demands and trends. Causal forecasting works on the assumption that future results are usually signaled by a particular variable or indicator. Each type of forecasting has its place within the business world. In this paper we will discuss the Delphi, Leading Indicators and Grass-roots models of forecasting.
The Delphi Method is a qualitative method of forecasting that uses the anonymity of the participants to give the input of each participant equal weight in the forecasting process. In the Delphi method a questionnaire or survey is issued to all participants. The participants in the Delphi method are usually from different areas of company operation. Each participant then responds to the survey submits their responses anonymously. The results are then compiled and sent back to the participants for further review and revision. The process is repeated until a common forecast consensus is achieved ( Goldfisher, 2003).
The Delphi method requires great communication among members of a company and depending on the size of the company and the amount of revision to the information submitted at each stage of the Delphi method can be a time consuming process. Both the Delphi an Grass Roots method require a great deal of communication between different facets of an organization in order for them to be successfully implemented.
Another qualitative method is the Grass Roots method. Grass Roots forecasting works on the assumption that the person closest to the customer or the person most involved with the end use of the product is able to predict the business trends of a product or service. (Chase, Jacobs & Aquilano, 2005) Grass Roots forecasting uses a bottom up approach in its preparation. The people closest to the customer compile their forecasts and submit their information to the next highest level, such as a distributor. The information is revised and passed up to higher levels until finally it is used as part of the decision making process in a firm’s business operations.
Grass Roots forecasting is similar to the Delphi method in that it requires a considerable amount of information to be compiled and communicated through company channels. It differs from Leading indicators forecasting in that LI
The leading indicators forecasting method is a causal method of forecasting. Leading Indicators forecasting works on the assumption that the behavior of one variable can be used to predict the behavior of another variable with a high degree of accuracy
( Lapide 2001). In LI a particular, statistic, situation or event is monitored. The monitored statistic is then used to predict the behavior of another event.
In my company we use several methods of forecasting depending on what aspect of company operations we are trying predict. However we primarily use the Delphi, Grass Roots and Leading Indicators forecast methods for our business operations.
We use the Delphi method to forecast operations that require a great deal of interdepartmental organization. The Delphi method allows us to pool the collective talents of individuals from different sections of the refinery to come to a consensus forecast. The Delphi method is most often used to forecast manpower and supplies for the upcoming quarter. Since these are resources that the refinery has to use collectively, it is imperative that the company use a forecast method that draws from personnel in all phases of its operation.
The anonymity of the Delphi method is particularly useful. The anonymity of the Delphi method allows us to come up with forecasts without the interdepartmental fighting that often occurs when resources have to be shared. By using the Delphi method participants do not know the identity of the other participants in the forecasting process. Thus they are less likely to generate forecasts that are intended to favor one group or aspect of company operations over another.
For the last three years the Delphi method has given exceptional results in forecasting manpower and supply requirements per quarter. It allows our organization to pool the collective knowledge of experts from different sections of company operations and arrive at n impartial forecast. One drawback to the Delphi method is that it is a time consuming process and can be negatively affected should a participant leave the company, go on vacation or be come sick
The Hovensa Refinery uses Leading Indicators forecasting to predict revenues for its winter quarters. Some of the refined oil products that we produce show a tendency of inelastic demand. For example heating oil has a demand curve that is very inelastic. During the winter months when the weather gets cold people will continue to heat their homes at a comfortable temperature, regardless of price. So for the fourth and first quarter of the year when heating oil demand is high, we use Leading Indicators forecasting to predict company profits for those two quarters.
The leading indicator that we use is the price of crude oil. Crude oil prices lead the prices of heating oil in the winter months by about three to four business days. So by keeping track of the price of crude oil, we’re able to forecast profits from the sale of heating oil very accurately. If the company sees a trend of high crude oil prices extending through the months of October through March then profits for the fourth and first quarter of the year will be high.
In the last four years the Grass-Roots method of forecasting has moved into greater prominence in my company. This is a very powerful forecasting method that I personally feel is overlooked and under utilized. Grass-Roots forecasting involves collecting data from the end-user, the person or persons who are most involved with the product just before it reaches the consumer, to forecast various aspects of business operations.
Recently we have been gathering input from our retailers, and regional distributors to forecast our production needs. Grass-Roots forecasting has made a significant impact in our company operations in that it has forced us to become a more responsive and nimble company. We now communicate with our end-users at least two to three times per week, and based on the input that they give to us we tailor our production planning to meet the Grass-Roots forecast recommendations.
GRF has allowed us to gain a better handle on the general trend of the way our products are utilized and demanded with varying market conditions. Also grass-roots forecasts give a significant insight into niches that affect our business. Many times the end-users have very detailed information about the area that they service, this information is first hand and very reliable. Our company has found that we have been able to more accurately tailor our products and services to our customers, and reduce waste.
For example last year one of our grass-roots forecasts predicted a slow down in the demand for gasoline in one of our Florida retail outlets. This was because the people in that area had on a large scale bought more fuel-efficient cars and had either sold or parked up all of their SUVs. We were able to scale back our deliveries of gasoline to that particular retailer and divert gasoline inventory to areas where the gasoline would more likely to be consumed. This sort of information and forecasting is not available from any other method except the grass-roots method.
GRF requires us to be more communicative from every aspect of our company and to its respond quickly to changes in market conditions. But the accuracy that I have observed from grass-roots forecasting is phenomenal and I believe that as the years progress this type of forecasting will work its way more and more into our company’s operations.
References
Chase R., Jacobs F., & Aquilano N., ( 2005). Operations Management for Competitive
Advantage 11th edition. New York: McGRaw-Hill.
Modified Delphi: A concept for new product forecastingGoldfisher, Ken. The Journal of Business Forecasting Methods & Systems. Flushing: Winter 1992-1993.Vol.11, Iss. 4; pg. 10, 2 pgs from Proquest Database
New developments in business forecastingLarry Lapide. The Journal of Business Forecasting Methods & Systems. Flushing: Fall 2001.Vol.20, Iss. 3; pg. 11, 2 pgs from proquest database
