Artificial intelligence approaches to achieve strategic control over project cash flows

The ability over the course of a construction project to make reliable predictions regarding cash flows enhances project cost management. This paper uses artificial intelligence (AI) approaches to predict cash flow trends for such projects in order to develop appropriate strategies that apply factors such as float, process execution time, construction rate and resource demand to project cash flow control. AI approaches involved in this paper include K-means clustering, genetic algorithm (GA), fuzzy logic (FL), and neural network (NN). K-means clustering is employed to categorize similar projects, while the other approaches are used to develop the Evolutionary Fuzzy Neural Inference Model (EFNIM), a knowledge learning model. FL and NN are employed in the EFNIM to develop a neural-fuzzy model that can deal with uncertainties and knowledge mapping. GA is used to optimize the membership functions of FL and NN parameters globally. The major target of this AI learning is to address sequential cash flow trends. This trained result is furthermore applied to a strategic project cash flow control. This cash flow control affects project performance within the banana envelope of the S-curve for project management.