Construction planning with machine learning

Machine learning is heavily hyped technology also in a construction context. It will over the next years be attempted to implement it many places in building processes. One particular place is during planning. As construction projects tend to be influenced by interrelated issues that result in cost and/or time overrun and lowered performance it has been attempted continually to develop planning and prediction methods that can mitigate these issues. This study aims at investigating the possible applications of machine learning in planning of construction projects and their impact on project performance?A literature review about machine learning (ML) in construction project planning is done. Methodologically two cases are selected. The first is drawing on a productivity survey of construction projects in Sweden. They are analysed to find the most influential factors behind project performance. The data encompasses project attributes, external factors, the project organization. Statistical correlation is used to find the features that are strongly correlated with four performance indicators: cost variance, time variance and client- and contractor satisfaction. A regression analysis is done to develop a model for predicting project cost, time and satisfaction. In the results, project technical complexity like the amount of prefabrication are among the features affecting project performance. Human related factors are of high impact and most likely to predict success including issues like the client role, the architect performance and collaboration. The second case conceptualise constructability and risk analysis in civil engineering projects. The development build on literature study, expert interview, unsupervised and supervised learning. Even more after the introduction of ML, there is a need for human reasoning in planning. It is not enough to include human aspects in the ML modelling, there is also a need for strengthening qualified  reasoning in the decision making in construction project planning.