Purpose
The purpose of this paper is to demonstrate that applied spatial modelling can inform the planning, delivery and evaluation of retail services, offering improvements over traditional retail impact assessment (RIA), especially within localities which experience seasonal fluctuations in demand.
Design/methodology/approach
The paper first describes a new theoretically informed tourist-based spatial interaction model (SIM) which has been custom-built and calibrated to capture the dynamics of the grocery sector in Cornwall, UK. It tests the power of the model to predict store performance for stores not used in the original calibration process, using client data for a new store development. The model is operationalised for the evaluation of various retail development schemes, demonstrating its contribution across a full suite of location decision making application areas.
Findings
The paper demonstrates that this highly disaggregate modelling framework can provide considerable insight into the local economic and social impacts of new store developments, rarely addressed in the retail location modelling literature.
Practical implications
Whilst SIMs have been widely used in retail location research by the private sector, the paper shows that such a model can have considerable value for public sector retail planning, a sector which seemed to have abandoned such models from the 1980s onwards, replacing them with often very limited and crude RIA.
Originality/value
The ability to review the forecasting capabilities of a model (termed post-investment review) are very rare in academic research. This paper offers new evidence that SIMs can support the RIA process.
