Generation of Solutions in Spatial Situational Analysis of Logistic Projects

In a geoinformation system (GIS), a logistics project functions as an informational entity that describes the spatiotemporal processes required for generation and maintenance of material flows. Given the incomplete description of the interactions between the logistics project and its external environment, the application of case-based reasoning (CBR) becomes appropriate. CBR enables the resolution of anomalies in logistics processes by leveraging past experiences. This study addresses the problem of developing spatial representations of solutions for anomalous situations represented as two-component images. Such an image represents extended knowledge about a set of precedents within a limited territory. Beyond the precedent scheme on a map, the proposed approach incorporates new cartographic objects in the image, reflecting possible, permissible, and impermissible transformations of the situation. These transformations should be viewed as invariants of the situation's essence. Use of images improves decision-making reliability. For solution generation, the following approach is proposed: a spatial image is constructed for a problematic situation and compared with images stored in the intelligent GIS. The solution image of the most semantically similar situation is used as the foundation for generating solution options. The generation process involves transforming the solution image to fit the spatial area of the anomalous situation and forming a set of acceptable instances based on semantic content. This paper elucidates the principle of spatial transformation of images and explores methods for selecting the optimal solution. The advantages and limitations of the proposed solution generation method are analyzed using data from the OpenStreetMap geoinformation service.