Research On Green Building Decision Optimization Based On Multi-attribute Decision-Making And Machine Learning

Resource depletion and environmental pollution are two huge challenges in today’s world.As the main living environment of human beings,architecture is one of the main contributors to energy consumption and pollution emissions.The construction industry must become the focus of energy saving and emission reduction,and at the same time should undertake the task of providing comfortable and healthy use space and good ecological benefits for human beings.China is currently the world’s largest emitter and largest energy consumer.To solve these two problems,it is imperative to promote the development of green buildings.Green buildings can achieve the goal of saving resources and protecting the environment as much as possible within the life cycle of the building,and provide people with a healthy and comfortable use space.The “Green Building Evaluation Standard” issued by the Ministry of Housing and Urban-Rural Development of the People ’s Republic of China is a rating system based on green building measures,which can provide third-party verification for green buildings.Uncertainty is an inherent feature related to sustainable construction project decisions.The main reason for this uncertainty is that the relative impact of stakeholders on their decision-making standards and the full impact of sustainable development decisions on construction cost input,construction service quality,and environmental benefits are not fully understood.At present,due to the large differences in the attributes of new construction projects and the uneven level of sustainable knowledge of green construction practitioners,it is challenging to make correct green construction decisions.At the same time,due to the vast territory of China,the natural economic and cultural differences among regions are significant,and general green construction It is difficult for the strategy selection strategy to be fully applied to the specific characteristics and needs of different regions in China.In this regard,the author conducted a study on the optimization method of green construction strategy for hot summer and warm winter regions,which helps green construction engineers The green construction technical plan makes rational and rapid decision-making to better promote the development of green buildings.One of the key decisions made to achieve the goals of a sustainable construction project is to choose green construction technical measures,so that the project meets the certification requirements of green building standards.This study proposes a multicriteria decision-making method for the selection of sustainable measures.A questionnaire survey of green construction experts in Shenzhen is conducted to understand the benefits(quality,ecology)and input of experts on the various measures in the green building evaluation standard.The feasibility evaluation in four dimensions(cost,technical difficulty)and the preference of green construction experts in Shenzhen for income input,based on the expert opinion,the MAUT method was used to establish a comprehensive evaluation model of measure items.Green construction measures are sorted to find the most feasible measure entries.The complete ranking of green construction measures can assist green construction engineers to make quick decisions.The complete ranking of the only and certain green construction measures obtained by the MAUT method cannot be directly applied to construction projects with different attributes.In view of the lack of practicality of the multi-criteria decisionmaking method,a machine learning algorithm is introduced,and each measure item under different project attributes Have different feasibility values,thus obtaining a complete ranking of specific green construction measures based on project attributes,using the green construction measure packages corresponding to different project attributes to build a database,and then using the database to train and fit the artificial neural network to let the nerve The network classifies all the national standard measures items into two categories,and identifies the selected items and the unselected items under the specific project attributes.And the actual project is used to verify the model,which proves the effectiveness and reliability of the method.This result shows that the developed method can provide valuable guidance in the process of selecting measures at the initial stage of project design.This study selects the technical strategy decision-making method of green buildings in Shenzhen area as the optimization object,and solves the problem of suitability for different green construction measures in hot summer and warm winter areas.Secondly,this study adopted a multi-criteria decision-making method,selected four conflicting and mutually restrictive attributes of the input and output dimensions as the evaluation system,and obtained a complete ranking of the green construction measures with excellent performance in the four dimensions,which can assist the green Construction engineers make decisions.Finally,the study uses machine learning algorithms to allow the computer to independently learn the green building experience,and quickly generate green building plans that meet the specific project attributes when encountering new projects,improve the quality of decision-making and optimize the suitability of green building plans.This paper’s research on green construction decision optimization methods can assist green construction engineers,especially new practitioners,to make quick decisions,while saving decision time and helping engineers minimize the negative impact of including measures in construction projects,and maximize Improve the positive benefits to the environment and building quality.