Predicting Trends

In a competitive environment, forecasting demand and being able to cater for increases in demand are crucial to the survival of an organization.

The increasing pressure on community service organizations to operate as corporate enterprises means they must be able to:

• Predict the future demand for their services
• Meet these demands through strategic planning processes.

Forecasting service demand is a useful tool in demand planning. It:

• Reduces uncertainties in the emerging requirements for services
• Allows for effective planning and provision of infrastructure
• Ensures the timely provision of resources to support the demand.

It is important to achieve an acceptable level of accuracy in forecasting to ensure confidence in the predictions and effective planning. This usually requires some skill and expert input.

Ideally, service demand forecasting should be established as a routine process within the planning cycle, with forecasts being updated whenever relevant data become available. The types of forecasting models include time-series and regression models.

Each type comprises:

• Demand averaging over a fixed period
• Trend analysis
• Seasonal influence estimate
• Cyclic variation
• Auto correlation estimate
• Irregular variations.

Computer packages allow more statistical forecasting. Packages include auto regressive and moving average processes.

In general, forecasting techniques may be categorized as either:

• Quantitative- statistical and other mathematical techniques which operate on accumulated data and extrapolated patterns of demand
• Subjective- used when no reliable data exist and when forecasts are made based on experience, intuition, judgment etc.

In reality, quantitative forecasts will also involve a level of subjective assessment, supported by rational argument.