Tapio Tyni

Evolutionary bi-objective optimisation in the elevator car routing problem

The paper introduces a genetic algorithms based elevator group control system utilising new approaches to multi-objective optimisation in a dynamically changing process control environment. The problem of controlling a group of elevators as well as the basic principles of the existing single-objective genetic elevator group control method are described. The foundations of the developed multi-objective approach, Evolutionary Standardised-Objective Weighted Aggregation Method, with a PI-controller operating as an interactive Decision Maker, are introduced. Their operation as a part of bi-objective genetic elevator group control is presented together with the performance results obtained from simulations concerning a high-rise office building. The results show that with this approach it is possible to regulate the service level of an elevator system, in terms of average passenger waiting time, so as to bring it to a desired level and to produce that service with minimum energy consumption. This has not been seen before in the elevator industry.

Elevator group control using distributed genetic algorithm

In this work we have studied the applicability of genetic algorithms to the optimization of elevator group control parameters. The goal was to minimize the average passenger waiting time in a simulated office building. The elevator group control consists of a set of elevators and controllers. At the highest level of the system hierarchy the elevator group controller decides which elevator serves which call. The decision is based on the actual calls, state of the elevators (location, direction, load) and on the estimation of the traffic situation like: incoming, interfloor, outgoing, which gives some idea how the calls will be distributed within a few minutes time period. Especially in office buildings the traffic type depends on the time of the day. This allocation problem, which seems to be quite simple, turns out to be a very difficult control problem in practise and is one of the features how elevator companies can competed with one another.

Evolutionary bi-objective controlled elevator group regulates passenger service level and minimises energy consumption

This paper introduces an elevator group control system based on bi-objective optimisation. The two conflicting objectives are passenger waiting times and energy consumption. Due to the response time requirements the powerful but computationally demanding Pareto-dominance based Evolutionary Multiobjective Optimisers cannot be used in this real-world-real-time control application. Instead, an evolutionary variant of the modest Weighted Aggregation method has been applied without prejudice. The presented approach solves the weight-scaling problem of the Weighted Aggregation method in dynamically changing environment. In addition, the method does not solve, but copes with the disability of the WA-method to reach the concave Pareto-front regions in the fitness space. A dedicated controller acts as a Decision Maker guiding the optimiser to produce solutions that fulfil the specified passenger waiting times over a longer period of time with minimum consumption of energy. Simulation results show that the control principle is able to regulate the service level of an elevator group and at the same time decrease the consumption of energy and system wearing.