Object detection for pattern analysis of consumer behaviour to optimise store layout

One of the crucial aspects of conventional retailing is the store layout as it is a critical aspect affecting consumer behaviour. This study explores the relationship between consumer behaviour and retail layout optimization with the help of data visualization through the use of close circuit video or CCTV gathered from online data repositories. This study uses computer vision model such as image recognition and tracking to measure the interaction between in-store customer with the store layout. This research assessed whether the model fit for the task by conducting series of test scenarios using data identical to what it would be in real-world business operation. In the scenario, the videos were divided into 120-seconds of snapshots to be processed to map the consumer interaction metrics and gauge how well the model fare to measure consumer behaviour. Post-simulation of the research shows that the tracking model used in the research is in-line with management theories and is usable to improve business activities, especially in analysing how effective the consumer-layout interaction works. The model also performs optimally and can operate in typical business situation as it able to provide insight that can be used for layout optimization.