Constrained spatial clustering method for facility location programming

Abstract

The invention discloses a constrained spatial clustering method for facility location programming. The constrained spatial clustering method includes describing spatial adjacent positions of target demand points according to triangle networks; counting, constraining and recognizing spatial clustering modes of the target demand points by the aid of side lengths of the Delaunay triangle networks from an integral level and a local level; providing the decision basis for spatial configuration of service source points. Spatial constraints such as spatial hindrances and convenience can be simultaneously taken into consideration when the spatial clustering modes of the target demand points are recognized, so that the practical spatial clustering modes of the target demand points can be acquired. The constrained spatial clustering method has the advantages that users do not need to formulate the quantities of the clustering modes of the target demand points, the spatial clustering modes under the consideration of the spatial hindrances and the convenience can be acquired from the integral level and the local level and can be ultimately visually outputted, and accordingly the important decision basis can be provided for facility location programming.

Description

technical field [0001] The invention belongs to the technical field of spatial data mining and spatial analysis, and relates to a constrained spatial clustering method oriented to facility location planning. Background technique [0002] Facility location is a special kind of location problem, which can be divided into two types: Weber type facility location and Paland type facility location. Weberian facility location refers to the problem of determining several service source points on the premise of known target demand points in Euclidean space. For example, given the spatial positions of several customers, determine the position of a warehouse so that the sum of the distances from all customers to the warehouse is the shortest. The demand of the Paland type facility location problem is planar distribution. On the premise of known demand distribution, the location problem of several service source points is determined. Spatial clustering methods are of great value for l...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the defects existing in the prior art and solving the technical problems existing in the application of spatial clustering technology to assist in solving the facility location problem, the present invention provides a constrained spatial clustering method oriented to facility location planning, which is a method that can simultaneously take into account Spatial barriers and convenience objects, spatial clustering methods that require less human intervention and can identify spatial aggregation patterns of different densities and shapes, further improve the practicability and reliability of spatial clustering technology to assist in solving facility location problems

Images