Application of artificial neural networks in predicting earthmoving machinery effectiveness ratios

Many constructional processes are carried out by machines working together and forming technological systems. An example here can be an earthmoving machinery set made up of excavators and means of transport. For process design purposes most important are the effectiveness ratios relating to the profits and losses stemming from system use, i.e. the system efficiency per unit work ratio W(N); the index of losses due to the idle times of the machines working in the system Sj; the output transport unit cost index Kj. This paper presents the results of applying neural networks in predicting effectiveness ratios, i.e. W(N), Sj and Kj for earthmoving machinery systems consisting of c excavators and N means of transport. It is showing the relevance to practitioners and researchers industry. The values of the characteristics can form a standard basis for designing construction earthworks. Having a dataset consisting of the technical parameters of earthmoving machinery systems and the corresponding effectiveness ratios one can train neural networks and then use the latter for the reliable prediction of W(N), Sj and Kj.