Abstract
This paper presents a new method for elevator supervisory control system in order to reduce long waiting passengers. In past, the methods of rules with AI or accurate arrival time prediction based on ETA(Estimated Arrival Time) have been applied, but they are not enough to reduce long waiting passengers. We propose a cars cooperative method for elevator supervisory control systems. The method consists of two phases: (i) prediction of generating long waiting passengers in fiiture, (ii) assigning a car to new hall call in consideration of present passengers and predicted future passengers. The result of experiment shows the proposed method can reduce long waiting passengers rate by 18%-51% from conventional one.
