Assessing wholesale market relocation decisions considering mobile vendor routing: A generalisable GIS-based spatial analytics model

Urban wholesale market relocations often impose additional travel burdens on mobile vendors operating at low margins in the informal sector and create spatial inequities when guided by non-spatial criteria. This study, aiming at assessing the impact of market relocations on mobile vendors, presents a novel GIS-based framework that combines the Clarke and Wright Savings Algorithm for route optimisation with Kernel Density Estimation, a machine learning technique, to identify high intensity vendor corridors. Heatmap values of vendor activity and average distance metrics were normalised and candidate market sites were ranked using a new Equity-Efficiency Index measure which accounts for landuse policy preferences. Candidate sites were mapped further considering gain/loss in accessibility, for classifying the service area into improvement and degradation zones weighted by population. A case study based on the relocation of Segiri market in Samarinda, Indonesia has been developed to illustrate the method. Results show that there is a new location in Samarinda that can deliver large efficiency gains in key corridors but resulting in widespread service area losses. Other potential locations achieve moderate, evenly distributed accessibility improvements. The framework delivers spatially explicit evidence of benefit and burden distribution and supports relocation decisions that balance logistical efficiency, equitable access and urban resilience.