The judgement of experts is both used directly as a method of demand or sales forecasting andalso to improve the reliability of quantitative techniques. It is used in the following ways:
Available fertilizer usage data on fertilizer types and provinces is provided to three orfour experts. The experts are asked to make estimates for the following season or year.A final estimate by consensus is reached at a meeting of the experts.
It may be difficult to find experts familiar with agriculture in every part of the country.For this reason, two experts could be selected for each province. For example, the headof the department of agriculture and the leading fertilizer distributor in each province canbe chosen. The estimates received from them are aggregated to form a range ofestimates. A national team representing, for example, the Ministry of Agriculture,provincial Departments of Agriculture, Agricultural Research Institutions, manufacturersand/or marketing organisations and the Extension Service then discuss the range ofestimates and arrive at a consensus.
The concept of using expert opinion for forecasting is known as the Delphi Method.Under this method the group's estimates are returned to the individual experts for reviewand a second round of forecasts is received from the experts. With each round thedegree of consensus improves.
Furnishing relevant background data to the experts improves the quality of the estimates as thisprovides them with a common base. If there are deficiencies in the base data, the experts areinvited to highlight them and appropriately qualify their forecasts.
The merit of the expert judgement method lies in the fact that it does not require data,elaborate statistical tools or expertise. The forecast can be made easily and speedily. Thisis a practical forecasting method when time series data over several years are not available.The disadvantage, however, is that it relies on judgement, hunch and intuition. This methodis not of much help in getting detailed break-down of demand figures by fertilizer types anddistricts. As the method depends on outside experts it is at once an advantage and adisadvantage. The advantage is that there is no bias in the forecast, as may be the case withforecasts made by district officials or field sales staff who may overstate or understate thedemand to suit their convenience and interest. The disadvantage lies in the possibility of theexperts, not taking the task seriously as they do not stand to benefit from the accuracy of theforecasts.
