Intelligent supervisory control for lifts: dynamic zoning

Abstract

It is generally accepted that a good lift system must provide the passengers with the best quantity, normally referring to handling capacity, and quality, normally referring to waiting time and/or journey time, of services with the most economical solution. To fulfil such an objective, the lift system must be adaptable to the everchanging traffic patterns of the building. Zoning has been one conventional way to achieve the goal. However, zoning is generally either fixed permanently or in other words, static, based on time scheduling where the floors being grouped into zones are predetermined during the design stage. In this paper, a new concept is introduced where zoning becomes a dynamic performance of a lift system, continuously changing with respect to the changing traffic patterns. The objectives are to achieve maximum handling capacity and minimum waiting/travelling time for passengers during up-peak, down peak and heavy interfloor traffic conditions. Two algorithms, namely ‘Uniform RTT’ and ‘Demand Matching’, have been developed for intelligent control of the dynamic zoning scheme. Comparison with a normal system without zoning is based on an available commercial software package using the same traffic patterns. Computer simulations on a pseudo building with 12 floors and three cars have revealed a significant improvement in the waiting time, journey time and handling capacity of passengers for most of the cases. The original concept of dynamic zoning was presented during an international lift conference.