Spatial data and information arises in many fields. Some of the areas that Data Analysis Australia has used spatial analysis are described include the following:
Data Analysis Australia's consultants understand spatial data using best practice statistical analysis including all standard geostatistcal and spatial points process methods. These methods can answer questions from "is this cluster of disease just due to chance?" through to "what is the best estimate of ore grades in this mine?".
In addition to this, we have up-to-date Geographic Information Systems (GIS) capabilities, ensuring that we make full use of maps as a powerful tool to visualise and communicate research results. Using such packages, our consultants are able to produce accurate and detailed map projections that overlay data from multiple sources.
Data Analysis Australia consultants' expertise in spatial analysis and mapping includes:
Examples of project experience where Data Analysis Australia has applied its spatial analysis capabilities are listed below.
Optimal Location Modelling of Metropolitan Magistrates Courts
Western Australia is geographically the largest State in Australia, and building and development is spreading both North and South along the coastal side of the Perth metropolitan area. Given these developments, and the increasing population, it is important to plan appropriately for the placement of new Court buildings. Data Analysis Australia created a Locational Model as part of the Metropolitan Courts Strategic Planning Project in December 2005. This work covered various scenarios aiming to optimise the placement of justice facilities by minimising travel from offence locations to court locations projected out to 2031. In 2006 the Locational Model was expanded to incorporate public transport. In 2007/08 the model was updated further, incorporating recent offence data and scenarios with location constraints reflecting the Government's strategic development planning for certain areas.
Optimal Location and Staging of Health Infrastructure
The health sector is characterised by the need of governments to have long term plans for their capital investment in infrastructure such as hospitals and health centres. This infrastructure must be appropriate in both design and location, to meet the needs of the community, both now and in the future. For the Health Department of Western Australia, Data Analysis Australia developed a comprehensive data driven model, that linked together information from a number of sources including the Department's morbidity database, Census information and the transport system, to provide a rational means of making decisions on the location and size of health facilities. This model took into account existing infrastructure capability and service coverage to inform future infrastructure development. A set of optimisation procedures was developed as part of the modelling process, to investigate the optimal staging of infrastructure developments. The model was in effect a geographical measure of demand for and access to health services.
Mapping of Population and Aged Care Facilities
In 2004 Data Analysis Australia produced a series of customised State and metropolitan population maps for the Silver Chain residential health service provider. The locations of Silver Chain's aged care infrastructure and services were superimposed upon these population maps allowing for assistance in the strategic planning of locations for future services.
Spatial Analysis of Childcare Services Demand
Socio-economic profiles of households across the City of Cockburn were prepared by Data Analysis Australia to inform the planning of community services. The project involved consolidating and displaying information from a number of data sources including Australian Bureau of Statistics Census data, through the CData software, and various data sources held by the City of Cockburn. The focus of this project was to gain an understanding of the spatial distribution of demand for childcare services across the City and to identify gaps in the geographic coverage of these services.
Analysis and Spatial Estimation of Tree Health
In 2008 Data Analysis Australia was contracted to undertake an analysis of data collected from the 2002 and 2008 surveys of Eucalyptus wandoo trees across the southwest of Western Australia. This botanical survey involved examining trees at approximately 130 sites along three large-scale transects covering most of the wandoo range. The data collected consisted of a number of variables including a composite measure of tree health (crown condition) in addition to environmental variables such as insect damage, tree density and fire history. The work completed involved a multivariate analysis of the relationship between crown condition and environmental variables including location. In addition to this, Geostatistical methods were used to model, estimate and subsequently map the crown condition of trees across the wandoo range in the southwest of Western Australia.
Mapping of Rebate Claim Dispersion
To encourage the uptake of products that reduce water consumption in Western Australia, the Water Corporation provided households with a range of 'Waterwise' rebates. In 2006 Data Analysis Australia was contracted to produce a series of maps to display the geographic dispersion of rebate claims for each product. An investigation of these maps indicated that rebates were more popular in the newer residential suburbs as opposed to well-established areas.
Mapping of Electricity Consumption
A series of detailed maps, in addition to summary tables, of annual average household electricity consumption were prepared for three Local Governments on behalf of Western Power. Each map was produced for two consecutive years and displayed the average consumption for each Collection District within the respective Local Government Areas. These maps highlighted geographic and temporal patterns in residential electricity usage.
Calculation and Plotting of Regional Population Projections
Data Analysis Australia maintains its own population projections model for the whole of Australia. In order to provide forecasts for small geographical areas that are of interest to our clients, the model allows forecasts to be created for any combination of Collection Districts within Australia. The key aspect that differentiates the Data Analysis Australia model from other forecasting models is the sophisticated statistical methods used in its internal migration component. In 2006 Data Analysis Australia was contracted by the City of Bunbury to forecast the population for a number of regions within Bunbury that were of interest to the City in its planning processes. We provided a base scenario and two additional scenarios that incorporated accelerated growth assumptions for the City.
Sampling Design and Implementation for a Travel Survey
The Perth And Regions Travel Survey (PARTS) was an ongoing study of day-to-day travel patterns conducted by Data Analysis Australia from 2002 to 2006 for the Department for Planning and Infrastructure. Data Analysis Australia developed sophisticated sampling methodologies to optimise the quality of data collected while minimising data collection problems through the use of modern geographical information system techniques. Data Analysis Australia designed all aspects of the spatial sampling methodology and implementation.