Publication

Abstract : Facility recommendation algorithm is used for multi-objective problems which are based on the transportation cost and accessibility for different applications. GPS-based devices are used to generate the trajectory of a road map which can serve as a major input for the right selection of geographical location for the various facilities. Major concern for the facility recommendation system is accessibility based on the distance between facility and access points. The ultimate aim of this paper is to analyze the facility recommendation system based on trajectory clustering using greater arc distance. Road trajectory data points are used in this analysis for better understanding. The number of clusters can be fixed based on two use cases: (1) the number of specific facilities to be planned in a city is known and (2) the number of facilities to be planned is unknown. For the first case, the number of centers or clusters can be directly given but, whereas for the second case, the number of locations for facilities can be calculated using the elbow method. The centroid point of the clusters formed can be recommended for the most essential facilities including primary health centers (PHC), schools, hospitals, ATMs, banks, and outlier points can be recommended for facilities like solid waste management (SWM) centers.