A Short Review of Blending Optimization Methods for Concrete Aggregates

"Optimization techniques are successfully used in many different fields such as engineering, physics, economy, and politics in order to enhance the performance of the systems. In concrete technology, Aggregates are prepared by mixing the water and cement in proportional quantities. In order to obtain desired quality control parameters, the determination of the ingredients ratios is one of the important problems for the optimization. Additionally, in the quarries where the aggregate quality constantly changes, blending optimization must be made to obtain a sustainable aggregate quality. This study showed that linear programming, which is easy and which includes many aggregate properties, cannot be used in terms of some aggregate properties. Therefore techniques such as nonlinear programming or genetic algorithm must be used. INTRODUCTION Concrete is a used construction material and made by mixing the aggregate, cement, and water. The concrete properties are directly related to the properties of these materials in blending. The quantity of aggregates in concrete is approximately 70 percent of the concrete’s volume. Therefore, the most important material is aggregates. The extraction of aggregate materials in quarries is more problematic because of variable bedrock properties. Therefore, the quality of the materials changes and the characteristics of the concrete are also influenced directly. In terms of sustainable use of natural resources, produced aggregate materials which have different qualities should be used optimally (Adiguzel et al., 2014). Nowadays, in the solution of the blending optimization, various optimization techniques are being used such as linear programming, nonlinear programming, and genetic algorithm. In this study, blending optimization techniques especially used for concrete aggregates blending have been summarized, and the advantages and disadvantages of these techniques have been introduced. BLENDING PROBLEMS Blending is a frequently encountered problem in process-type manufacturing such as mining, refinery, food industry, chemical industry and metal industry (Sakalli 2010). Blending is a commonly used method in lots of practical solutions (Sarker and Newton 2008). Several optimization techniques are available for blending problems (Furman and Pardalos 2009). The most important issue is to blend with minimum cost. The blending problem is a special linear model which was developed to solve this problem. In blending optimization research, linear programming is a widely used method (Adiguzel et al. 2011; Bascetin and Adiguzel 2011)."