Optimization on Ready-Mixed Concrete Dispatching Problem via Sliding Time Window Searching

Abstract

The dispatching of ready-mixed concrete (RMC) is usually scheduled manually and difficult to meet the demand of construction sites. The inefficient dispatching plan causes concrete defects of discontinuous casting (cold joint formation) or long on-site queuing (concrete early hardening inside RMC truck). The RMC dispatching literature focus on optimizing the dispatching intervals to minimize RMC truck on-site queuing, system operation costs of RMC plant and construction sites, or tradeoff. The previous research methods include mixed-integer linear programming (MILP) and evolutionary algorithms. They have the limitations of providing optimal solutions or computation efficiency. This paper develops a Binary Linear programming with Sliding Time-Window (BLSTW) to overcome these limitations. The BLSTW model subtly determines the dispatch of RMC trucks through binary variables incorporated into sliding time window search approach. Two perspectives, from the RMC plant supplier and construction site demanders, are discussed respectively. Three case studies are conducted to verify the efficiency improvement of the developed model including one largescale example of 25 sites. The comparisons made with two cited results shows the superior of the developed model.