Genetic Algorithms Application and Testing for Equipment Selection

• Published in: Journal of construction engineering and management Vol. 125; no. 1; pp. 32 - 38
• Main Authors: Haidar, A, Naoum, S, Howes, R, Tah, J
• Format: Journal Article
• Language: English
• Published: Reston, VA American Society of Civil Engineers 01.01.1999
• Subjects: Applied sciences; Buildings. Public works; Computation methods. Tables. Charts; Construction equipments; Construction works; Exact sciences and technology; Structural analysis. Stresses; TECHNICAL PAPERS; System architecture; Evaluation; Decision support system; Test in use; Knowledge base; Case study; Hauling equipment; Excavating; Genetic algorithm; Construction equipment; Mine; Selection criterion; Numerical simulation; Cost analysis; Application
• ISSN: 0733-9364; 1943-7862
• DOI: 10.1061/(ASCE)0733-9364(1999)125:1(32)

This paper describes a research undertaken at South Bank University that investigated the feasibility of applying artificial intelligence methodologies to the optimization of excavating and haulage operations and the utilization of equipment in opencast mining. The selection of the excavating and haulage equipment to remove the overburden in opencast mining has a significant effect on the profitability of the operation. Thirty-five to fifty percent of the total cost of operating an opencast mine is attributed to excavation costs and 15-20% of the total cost is attributed to haulage costs. The decision to select equipment is often based on past experience, location, and different organizational pressures, as well as complex numerical computations. Therefore, the research was directed into the development of a decision support system XpertRule for the selection of opencast mine equipment (XSOME), which was designed using a hybrid knowledge-base system and genetic algorithms. The knowledge base within XSOME is a decision-making task utilizing a decision tree that represents several nested production rules. The knowledge base relates mainly to the selection of equipment in broad categories. XSOME also applies advanced genetic algorithms search techniques to find the input variables that can achieve the optimal cost. The system was tested on four case studies to validate its accuracy. For each case study the equipment selected by XSOME was analyzed and compared with the actual equipment used by the contractor in the mine. A sensitivity analysis was performed on each case study to provide potential suggestions in areas where improvements could be made.