Master’s Students to present at Canadian Association of Geographers Conference
Two members of our spatial analysis lab will presenting their Masters work at the upcoming Canadian Association of Geographers annual meeting. They have been accepted to present in one of two GIScience special sessions chaired by Dr. Tarmo K. Remmel of York University and sponsored by the CAG GIScience Study Group and ESRI Canada Education and Research Group. We hope to see you at the conference.
Estimating Market Potential Using Census Data
Andrei Balulescu, Derek T. Robinson, M. Bogdan Caradima
Understanding whether a consumer base is available for a retailer can make the difference between business success or failure. As described in literature, retailers often use ‘rules of thumb’ in making locational decisions for store expansion. Four methods are developed to estimate retail expenditures in Ontario using census data, each systematically incorporating additional information. These are compared to provincial sales data for accuracy assessment. The methods are applied across three geographic census levels and the distribution, patterns, and casual effects of expenditure estimates are described. Regression and spatial statistics are used in a GIS environment to create spatial profiles of the consumer base. When applied across the landscape, prime regions for retail expansion are identified. Our results outline key variables underlying the location choices of retailers and describe the spatial pattern of expenditures across Ontario.
Criteria development for a suitability analysis of retail development across Ontario, Canada
M Bogdan Caradima, Derek T Robinson, Andrei M Balulescu
University of Waterloo, Geography and Environmental Management
As part of a study implementing a suitability analysis of retail development across the entire province of Ontario, Canada, criteria were selected and calculated using extensive script automation and large data sets. Suitability criteria at the site and situation level were generated as attributes for approximately 4.7 million parcels in Ontario, including topographic statistics, land cover proportions, and network distance from highway ramps. Across Ontario, parcel characteristics such as elevation, slope, and land cover will be used to determine parcel characteristics and development cost. Situation-level criteria such as the network distance from highway ramps provide a representation of accessibility and visibility of a parcel to high-traffic intersections. Challenges in calculating criteria using large data sets with limited computational resources include data accuracy and completeness as well as the conceptual representations of generalized spatial data. The presentation will highlight some of the challenges in generating criteria for a province of over a million square kilometers and present some of the solutions used to overcome these challenges.