Shooting-inspired fuzzy logic expert system for ready-mixed concrete plants

摘要

With a goal of reducing the overall production process variability of Ready-mixed concrete (RMC) and decreasing production costs, a fuzzy logic-based expert system was developed. Inspired by shooting sports, the system guides the plant chief technician by recommending the actions to be taken to reduce variability. Such guidance is based on the evaluation of the compressive strength and coefficient of variation results from the samples molded by those responsible for testing concrete. The system was tested in a RMC plant with monthly production of ≈ 3300 m~3. The results indicate that no major interventions are necessary for the investigated plant; however, training is likely recommended for some of the staff so as to attain a lower overall variability. These conclusions are in agreement with the technician's interpretation, proving that the system is satisfactory for application in the process. Additionally, a cost analysis was performed, and the results show that a potential reduction of 5.6% in the monthly production cost can be reached. This represents an attractive reduction since the RMC profit margin is about 8.0%.