11 th International Research/Expert Conference ”Trends in the Development of Machinery and Associated Technology” TMT 2007, Hammamet, Tunisia, 05-09 September, 2007. STRATEGIES FOR CONTROLLING ELEVATOR GROUP SYSTEMS Berna Bolat YTU. Faculty of Mech. Eng. Mech. Eng. Department, Besiktas, 34349 Istanbul Turkey İlyas İstif YTU. Faculty of Mech. Eng. Mech. Eng. Department, Besiktas, 34349 Istanbul Turkey Mustafa Alişverişçi YTU. Faculty of Mech. Eng. Mech. Eng. Department, Besiktas, 34349 Istanbul Turkey ABSTRACT In high-rise buildings, elevators must be safely and efficiently moving people during daytime. Generally, elevator up-peak traffic is predominant traffic between a lobby and various upper floors. Especially, the main expectation from elevators in buildings is increasing in a rush-hour time. Therefore, the passenger’s waiting time in their floor are increased. We can say that the services given by the elevator group control system are not sufficient in this building. The passenger service level in an elevator system depends on the controlling of the elevators. The purpose of this paper is to demonstrate that an effective way to control of elevators group using optimization algorithms and the results show that the average journey and waiting time of passengers are decreased by means of suggested method. Keywords: Elevator, elevator traffic, vertical transportation, up-peak traffic. 1. INTRODUCTION Elevator has been used safely and efficiently moving people in buildings for 150 years. If the elevator did not exist, buildings would be limited to five or six stores and also most of the high-rise buildings of the 20th and 21st century would be impossible. The number of high-rise buildings over 20 floors has increased rapidly in all over the world. Therefore, elevator is the most used vertical transport systems in the world. By creatively applying new technologies, elevator system is continued to develop for ride quality, reliability and safety in high-rise buildings, besides, to give better service to passengers, providing service at every floor in the buildings to as many as possible at a minimum time. In other words, an optimal elevator system must be reduced passenger’s journey, travel and waiting time in a rush-hour time. In a multi-stores office building, passengers take the elevators to the lobby floor forming the up-peak traffic at the beginning of the day. Hence, elevator traffic can be calculated analytically for up-peak situations where passengers arrive at the entrance floor and travel to the upper floors. Although the other traffic situations exist, such as outgoing, two-way or mixed lunch-hour traffic, the elevator group control strongly affects the up-peak traffic [1]. The first aim of this study is to describe optimization method of a group elevator that optimizes passenger service level. Therefore, the genetic algorithm in this study is applied to tackle the up-peak traffic problems in high-rise buildings and the optimization results obtained from the algorithm has been discussed. 775