Spatial behaviour in the retail environment
Publication Date
1991
Creator
Abstract
This research has focused on the development of techniques for the collection, analysis and presentation of large movement and behavioural data sets. The research has followed three strands: the investigation of pedestrian flow networks in the central business district, the relationship of pedestrian flow and retail turnover, and the study of movement and behaviour of customers in-stores. This thesis reports on the development of a self contained time-lapse camera system. The cameras were used to record flow conditions and people in both the city centre and in-store environments. For each person seen in the films a number of demographic and behavioural variables were extracted. These were then used as the data base for computer modelling systems. The investigation of the potential development of a turnover interface was necessary to test whether the variables extracted from the film were of any predictive value. It was found that it was possible to segregate the shopping from the non-shopping population on the basis of the data collected. Two movement models were developed. The first, an Origin- Destination model called WONKA, was based on the network estimation procedures used in vehicular modelling. WONKA was applied in both environments for the prediction of paths through the studied network. PRETTY, a customer simulation and animation program, recreated the recorded data and allowed the user to select data sets according to any combination of the recorded variables. Subsequently a probable path module was added to PRETTY as it became apparent that the tight theoretical basis of WONKA was not suited to the free layouts of many stores.
Item Type
ethesis
Thesis Type
PhD
Subjects (LC)
Associated Schools / Departments
School of Geography
eprints ID
12594
UoN Repository URI
Except where otherwise noted, this item's license is described as
File(s)![Thumbnail Image]()
Name
314444.pdf
Type
Full-text
Size
35.16 MB
Format
Adobe PDF
Checksum ()