Using machine learning to estimate the traffic mix and intensity in a building

It has long been believed that the number of up stops and down stops in a building, as well as the ratio between them, could be used to estimate the mix of traffic prevailing in the building and its intensity. With modern lift traffic analysis and data collection methods, it is now possible to generate large amounts of representative data in a reasonable time and with reasonable processing power. This paper attempts to use the generated data as training and testing data for a machine learning application that could estimate the mix of traffic as well as the intensity of traffic in a building. The method will first be applied to one or more representative buildings and then extended to more general cases.