A Mixed-Integer Nonlinear Programming Model for Minimising Construction Site Noise Levels through Site Layout Optimisation

Activities undertaken on a construction site are often accompanied with high levels of noise. Addressing the issue of noise pollution in construction is gaining significance with the growing awareness about the social and environmental components of sustainable construction and the increasing numbers of projects being undertaken in congested urban areas. The documented methods for reducing noise pollution in construction include controlling (1) the noise produced at the source; (2) noise levels reaching a receptor; (3) noise propagated along the transmission path. Methods addressing the latter points use the fact that attenuation of noise increases as the transmission path gets longer. Thus the efficiency of such methods can be improved considerably through optimising the arrangement of temporary facilities on construction sites, with respect to a receptor, making use of noise attenuation due to distancing noisy facilities away from noise-sensitive receivers. The building under construction can also be used as a barrier to the noise transmission path, where obstruction of particular facilities from a given receiver can help in producing lower levels of sound as measured at the receptor. The available literature on site layout planning is extensive but limited to only achieving traditional construction project objectives (travel and material handling cost, safety, etc.). This paper presents a mixed integer non-linear programming (MINLP) model that optimises the location of temporary facilities on site in order to minimise the sound levels measured at a pre-defined receptor. The present model is expressed in three stages: (1) defining the noise objective function; (2) implementing model constraints; and (3) application of COUENNE to solve the MINLP for a case study.