Совместное применение пакетно-узлового метода и технологий искусственного интеллекта в календарно-сетевом планировании строительных проектов

Introduction. During a series of experiments on compressing the schedule of the calendar-network schedule for a construction project, a hypothesis was put forward and confirmed that the use of Advanced Work Packaging (AWP) construction method allows for a 25 % reduction in planned construction time without changing the duration of the operations performed or allocating additional resources. However, although this approach solves the problem of schedule compression on the one hand, on the other hand, it creates additional problems or tasks related to an increase in information flows and planning volumes. Current achievements in the field of digitalization allow us to put forward a new scientific hypothesis that the problem of high-level detail in planning data can be solved using artificial intelligence technology. The goal of the study is to reduce the time required for construction and installation work through detailed planning and high-quality project implementation according to a scenario referred to as “Scenario – AI”. To this end, it is proposed to move the start of detailed planning to the initial stage, known as “Business Planning and Life Cycle Modeling”. Achieving this goal involves the use of machine learning (ML) and artificial intelligence (AI) technologies, which are very relevant in modern construction management. Materials and methods. The process model Planning & Scheduling was adopted as the initial methodology of calendar-network planning. The study examines different scenarios of planning and implementation of a construction project: one uses data with a level of detail increasing as it is received, and the other scenario uses the Batch-node construction method. Results. The conducted research and experiments confirmed the relevance of integrating machine learning (ML) and artificial intelligence (AI) into the construction industry. This area is of significant scientific and practical interest, opening up new horizons for improving the productivity, safety and quality of construction projects. The authors analyzed neural network technologies for solving time management problems and other calendar and network planning problems. The final solution to the research problem was the initiation of the project “Development and implementation of the AI Planner and AI Assistant Project Manager system”. Conclusions. The study concluded that the integration of ML and AI into the construction industry is an important step towards sustainable development of the industry, which is confirmed by the results of the study and requires further study to confirm and develop hypotheses.