Neural networks for cost estimating in project management

The purpose of this thesis is to evaluate whether neural networks can be used for cost estimating in project management and if so, whether they lead to improved estimates. Chapter one gives a short introduction into the field of artificial intelligence and describes the structure and learning algorithm of a selection of some typical neural networks. Chapter two reviews literature regarding cost estimating in project management and differentiates between different methods of cost estimates. Chapter three describes the statistical problem of parametric cost estimates and addresses these problems with the help of neural networks. Chapter four and Chapter five give examples of how neural networks can be used to represent mathematical equations which are currently used for cost estimating in project management. Chapter six applies neural networks on real world data and compares its performance with one of the leading parametric cost estimating software tools. Chapter seven includes the discussion and conclusions of the findings throughout the thesis. The limitation and restriction of the implementation of neural networks are examined and the potentials for further research are suggested.