Estimating the elevator traffic system arrival rate using exponentially weighted moving average(EWMA)

The passenger arrival rate is a fundamental term in quantifying the demand in elevator traffic system. The peak value of the arrival rate is the basis for determining the number and the specification of the required elevators. The ability to estimate the elevator passenger arrival rate in real time enhances the controller's decisions such as selecting the suitable number of sectors and switching between algorithms suitable for different traffic modes (e.g., incoming, outgoing, inter-floor). This paper discusses the use of exponentially weighted moving average to estimate the arrival rate of passengers in real time. It also studies the optimum value for the smoothing constant that minimizes the sum of the square of the deviations of the estimated arrival rate from the actual arrival rate. It is shown that the exponentially weighted moving average (EWMA) provides a better estimate of the arrival rate than the simple moving average (SMA). The arrival rate estimation using EWMA was also applied to real life passenger arrival data for two buildings.