Spatial Analysis on Networks: Towards identifying similarity of routes

Abstract

Finding an optimal path in a road network is a method of planning and decision-making that is mainly related to transportations and emergency response. The paper presents an algorithm for finding optimal paths in spatial networks, through the utilization of open source GIS and mathematical analysis of Networks using Graph Theory as well as using geographical proximity attributes of network nodes. The geometric and spatial information of the network as well as its relations with points of interest (POI) of the study areas located at the nodes and edges of the network, are transformed into spatial information, which by applying spatial queries in a geographical database (Postgis/Pgrouting) give query-enabled paths. The case study for the application of the algorithm and finding a route based on spatial queries is the island of Lesvos. This island combines intense topography and a complex road network with multiple geometric relationships. The area also has points of interest such as cultural, tourist and social. The final route will be a synthesis of morphological, tourist and cultural elements similar to the spatial search queries. Finally, the methodology as well as the search algorithm can be applied to any Spatial Network (transportations, environment, energy) described by its geographical features, considering all kinds of geographical issues, thus solving spatial problems and contributing to local development.

Towards a similarity index of network paths in spatial networks

“…For the spatial network analysis, we used the methodology and search algorithm proposed in the article by Agourogiannis et al, 2021. This process involves preparing the network based on graph theory (creating a spatial network with nodes and edges), inserting geographic information at nodes and edges such as distance, altitude and distance from points of interest, and finally searching for paths with spatial criteria.…”

“…The aim of this paper is to quantify the spatial relevance between paths in spatial networks by using a new multivariate spatial similarity index. The index compares the geographical properties of paths derived using a methodology and algorithm for path search in spatial networks, as shown in previous work (Agourogiannis et al, 2021), and calculates the similarity between the values of the solution paths and the target path. The quantification of similarity in geographical aspects is quite novel especially in terms of multivariate, aggregate and weighted similarity.…”