Copyright © 2009
Data Analysis Australia

Forecasting

Forecasting is an essential part of management. Predicting, with reasonable certainty, what might happen and how likely it is to occur is the first step in making informed decisions.

 

Forecasts are usually based on understanding trends and patterns in historical data. Statistical models are central to this. A distinguishing feature of Data Analysis Australia's approach is the use of structural information from the system being analysed. Rather than using black boxes, models are developed where the structure reflects reality and actual processes.

 

Data Analysis Australia also incorporates subjective factors into forecasting models. Structured consultation with expert panels and stakeholders is used to develop critical probability points in the models. An advantage of this approach is that it gives measures of how uncertain the future may be.

 

Some of the forecasting analysis Data Analysis Australia has completed include:

  • Demand for air services that contributed to the development of the master plan for the Christmas Island airport. This study required the assessment and modelling of major but uncertain infrastructure developments. Risk analysis methods were used to establish immediate needs while allowing a master plan to be developed that allowed for possibly higher demands in the longer term.
    • Demand for justice facilities such as courts in Australian cities.  Typically the need is to forecast the need for courtrooms, hearing rooms and even holding cell capacities.  While it is possible to readily identify some drivers such as demographic trends, there is also a need to understand how the raw demand translates into actual facilities.  Here our structural model approach was critical.

     

    • Future electricity consumption for a major Western Australian utility. Forecasts are generated not only for total consumption but also for peak demand since this directly affects generation capacity and capital requirements.  Medium term models for scheduling and short term models for energy trading have been developed.
    • Daily water consumption in Perth and measuring the extent to which it is influenced by the weather and advertising of conservation measures.  As Australia faces a possibly drier future, water management is becoming more critical and such forecasts are the key to both short term planning or restrictions and the long term planning of resources.