Artificiell intelligens som beslutsfattande medel vid ROT-arbeten : En kvalitativ intervjustudie för kartläggning av beslutfattningsprocessen

Sammanfattning: The construction industry stands for a third of the total carbon dioxide emissions globally , this includes the building process as well as the usage of the buildings. This in combination with the need for renovating older buildings in Sweden can lead to additional stress on the environment. This project aims to map out the decision-making process for property owners and consultants when working with renovation projects. The purpose of the study is to identify problematic areas that hinder efficient decision-making and sustainability efforts. With the problem areas mapped out, different solutions containing artificial intelligence will be explored and discussed. Ethical implications with the implementation of artificial intelligence will also be discussed in this study. In this qualitative study, a set of steps were taken to answer the research questions. Firstly, a literature review was conducted to explore existing research. Secondly, semi-structured interviews were held to gather empirical data. Lastly, the interviews were transcribed and analyzed with thematic analysis, to identify problematic areas in the renovation projects. The research strategy applied for this study is abductive. Due to the existing pandemic, the usage of software such as Zoom and Microsoft Teams have been used to conduct the interviews. The results show that the factors hindering the decision-making process were regulations & laws, economy & sustainability, lack of recycling, and lack of documentation. These were areas that both the consultants and property owners described as bottlenecks in the decision- making process. Artificial intelligence solutions were discussed for the problem areas regulations & laws and economy & sustainability because there is statistical data that could be used for training an artificial intelligence, unlike the other two problem areas that have to be dealt with manually due to the neglect of workers. The artificial intelligence solutions presented in this study can be considered ethical due to their assisting purpose. Although this research provides ideas for the implementation of artificial intelligence they are very brief and theoretical, further development and exploration are needed to implement the solutions.