Intensive urbanization and the transformation of historic urban environments require broader approaches to the digital documentation of architectural heritage. Existing digital platforms predominantly document individual architectural objects, while insufficiently representing their relationships with the natural, spatial, and sociocultural context. This study develops a conceptual model for the digital documentation of architectural heritage based on the Historic Urban Landscape (HUL) approach, using Almaty, Kazakhstan, as a case study. The research included a survey of 200 residents, an analysis of the ArchiCode database and four tourism-related online resources, adaptation of HUL principles, and expert evaluation of the initial model. The results showed that, within the study sample, perceptions of Almaty encompassed not only architectural objects but also the natural landscape, green and water infrastructure, public spaces, and place-related memory. The proposed model comprises a basic level documenting the architectural object and a HUL-based contextual level representing its relationships with the historic urban landscape. Expert evaluation of the initial model by 12 experts yielded Aiken’s V values of 0.94–0.96. The model was subsequently refined based on the experts’ recommendations. Further research is required to test its practical applicability in digital environments and other historic cities.
3 October 2024 in Herald of the Kazakh-British technical university
Effective management of financial resources in projects is crucial for project success. Often, difficulties with financial resources, such as budget overruns, lead to unfavorable consequences that directly impact the successful completion of the project, the quality of the outcome, and stakeholder satisfaction. Therefore, identifying and developing tools for the effective management of financial resources, is of paramount importance. The purpose of this study is to apply non-traditional earned value management (EVM) models based on machine learning to predict project costs. To achieve the research objectives, previous literature on the topic was analyzed, a dataset of past projects was prepared, and a machine learning model was applied. The study found that non-traditional models, such as the regression algorithm AdaBoost, produced results close to the actual costs. The research indicates that the developed model could become an indispensable tool for project management and business decision-making, as it demonstrates the ability to adapt to various conditions and make accurate forecasts.