Site selection is one of the most important factors in determining business competitiveness and performance. This is particularly true for a petrol filling station business, which relies on customer visits. In absence of the exact information on the number of customer visits, traffic volume can be treated as a proxy to it. This paper uses a 2Ω_D surface rendered using an multiple regression model and geographic information system (GIS) to identify site potentiality for petrol filling station business within residential neighbourhoods. In so doing, the paper has three objectives. Firstly, to briefly review situs characteristics that influence the level of traffic volume passing through a particular site; secondly, to specify a regression model that produces parametric estimates of these characteristics for predictive purpose; and finally to incorporate the model into a geographic information system (GIS) to render a 2Ω_D contour map of predicted traffic volume counts in a study area to assess site potentialiy for petrol stations. The study area is locted within the Central Johor Bahru Municipality (MPJBT) that covers the Mukim of Pulai and a part of Mukim of Senai_Kulai. Based on the model developed, it is found that type of artery road has greatest influence on traffic volume followed by accessibility, direction, site location and population density. A GIS map of predicted site potentiality for petrol station based on traffic volume has also been enerated. This study has illustrated how MRA and GIS can be used in examining site potentiality based n traffic volume counts for petrol filling station business.