A Machine Learning Approach on Earthmoving Fleet Selection

Abstract

This research underlines the importance of simulation modeling by presenting a fleet selection optimization methodology featuring a detailed simulation of earthmoving projects and a genetic optimization algorithm. Two genetic algorithm techniques for fleet selection are proposed and compared in a load and carry application. The simulation model takes into account specific cost criteria that go into equipment selection such as the ownership and operating costs of each machine. The impact of different scenarios on the Profit & Loss function (fitness value) and machine selection is extensively detailed and analyzed. The research work underlines topics such as machine size class and maintenance and repair. The methodology is applied to a real case study that took place in Lebanon. The proposed approach is shown to yield a significant increase in the efficiency of the selected construction equipment fleets.