Data-Driven Simulation-Based Analytics for Heavy Equipment Life-Cycle Cost

Abstract

Heavy civil and mining construction industries rely greatly on the usage of heavy equipment. Managing a heavy-equipment fleet in a cost-efficient manner is key for long-term profitability. To ensure the cost-efficiency of equipment management, practitioners are required to accurately quantify the equipment life-cycle cost, instead of merely depending on the empirical method. This study proposes a data-driven, simulation-based analytics to quantify the life-cycle cost of heavy equipment, incorporating both maintenance and ownership costs. In the proposed methodology, the K-means clustering and expectation-maximization (EM) algorithms are applied for input modeling to distinguish the maintenance stages, and to further generate corresponding distributions of these points. These distributions then are used to quantify the uncertainties embedded in the equipment costs through simulations. A historical data set of ownership and maintenance costs for a mining truck model was used to demonstrate the feasibility and validity of the proposed approach. This approach was proven to be effective in predicting the cumulative total cost of equipment, which provides analytical decision support for equipment-management practitioners.