evaluation of concrete plants readiness to produce high quality concrete for municipal constructions using past information

The only way to test the ability of concrete plants to produce high quality concrete is to test their final products. also, the process of testing and controlling concrete quality is time consuming and expensive. in this regard, having a quick, cheap and efficient way to predict the readiness of concrete plants to produce high quality concrete is very valuable. in this paper, a probabilistic multi-attribute algorithm has been developed to address this problem. in this algorithm, the goal is to evaluate readiness of concrete plants to produce high quality concrete based on the error rate of concrete compressive strength. using past information and data mining techniques, this algorithm predicts the readiness level of concrete plants by similarity of their production factors to past information. readiness alternatives for plants are ranked using data mining techniques for order preference based on their production factors (pf) and by evaluating the similarity/difference of each pf to past information. a case study of 20 concrete plants is used to illustrate the capability of the new algorithm; with results showing that the algorithm generated nondominated solutions can assist plant managers to set efficient production plan, a task both difficult, cost and time-consuming using current methods. in the case study, lab test totally confirm the algorithm outcomes thus it has been successfully verified.