Stock aggregation and visualisation in the NSW residential stock

Stock aggregation and visualisation in the NSW residential stock

Authors:

Daniel Daly, Georgios Kokogiannakis, Pratibha Acharya, Jun Ma, Jennifer Wood, Pascal Perez, Paul Cooper and Despina Clancy

Organisation of Presenter:

University of Wollongong, Australia

Abstract:

This paper will present the results of two recent projects focussed on improving the understanding of the NSW residential building stock: the NSW Housing Stock Mapping Pilot Project and the NSW Typology Development Project. These project have been undertaken with significant collaboration between staff from the Office of Environment and Heritage (and other State Government departments) and the SMART Infrastructure Facility and Sustainable Building Research Centre at the University of Wollongong. The overall purpose of the collaborative activities was to understand and map the potential market for and impact of sustainable housing products and sustainable building methods for the existing NSW building stock.

There was significant interaction between the two projects, as both the Housing Stock Mapping and Typology projects were aimed at transforming relevant data stored in separate, and often inaccessible, databases into a resource that can provide reliable and useable insights to a range of stakeholders. The two projects therefore both had a need for quality input data. The first stage of both projects was to review the existing data on the NSW housing stock, and to evaluate the availability and accessibility of data relating to important building attributes. Issues of data quality, in terms of collection accuracy, inconsistency in records, and database format, were an important consideration in the initial stages of the projects.

The Housing Stock Mapping project then compiled the datasets identified for inclusion into a central data warehouse, and defined consistent data definitions to fuse the disparate data sources. Example, proof-of-concept visualisations were then developed to demonstrate the analytical potential of the fused databases. The typologies project used a hybrid approach to define representative dwellings. The background data was evaluated to be insufficient to take a data-driven approach to typology definition; therefore the available data was augmented with a qualitative approach to garner expert knowledge regarding the most common housing types in the NSW stock. Eight broad residential typologies were defined suing survey results and an expert workshop.