Effect of batch size in repetitive construction

Most research related to repetitive construction assumes that all activities have an identical construction unit. But in reality, different activities may have different construction units, such as the construction of a block of townhouses consisting of several units sharing certain structures. The operations of pile driving and roof frame of all houses in a block are performed all together in one time whereas certain activities, i.e. painting and installing sanitary wares, are performed one house at a time.Therefore, construction plans derived from the traditional planning method may not be consistent with the actual operations.  This article discussed the construction scenarios when different activities had different batch sizes.  It emphasized the impacts of batch size of construction activity on batch throughput time, project duration, and batch in production, when all activities had same production rate and the relationships between activities were Finish-to-Start (FTS).  Given an example network, two cases were compared, activities with an equal batch size and those with different batch sizes.  The result showed that, in case of equal batch size, processes with smaller batch sizes had shorter batch throughput time, project duration, and batch in production. This is because, with smaller batch sizes, the cycle time of each operation reduced and activities could start their work earlier. In addition, batch in production (or work in process) was found only processed time with no product waiting between processes. Reducing batch sizes of certain activities leaded to a process with different batch sizes. It was found that, in this case, batch throughput time, project duration, and batch in production were dependent on not only the batch size but also the position of activities of which batch sizes were reduced. That is, only the case where reducing the batch sizes of at least two consecutive activities would have beneficial result. Moreover, batch in production (or work in process) was products in process and product waiting between processes, the later occurred when two consecutive activities had different batch sizes. Lastly, this paper proposed a method for estimating batch throughput time, project duration and batch in production, and this can be part of process design in reducing waste within construction operations.