Estimation of earthworks execution time cost by means of artificial neural networks

The paper presents a methodology for determining earthworks execution time and cost. The accurate determination of the productivity of machines working in sets is a major problem in earthworks execution planning. Earthworks are conducted in random condition and so the parameter values achieved during their realization have a random character. Using an existing database based on measurements carried out on a construction site, an artificial neural network was built. The latter was used to predict productivity for selected sets of machines and to calculate the task execution time and cost. The obtained set of results allows one to select the machinery set optimum with regard to earthworks execution cost or time.