Genetic Network Programming Research Papers

Elevator group supervisory control system using GNP with reinforcement learning based on normalized information

2025

Genetic Network Programming (GNP) has been proposed as a new method of evolutionary computations and its basic characteristics are studied. Recently, GNP is also applied to Elevator Group Supervisory Control System (EGSCS) which is a... more

descriptionView Paper arrow_downwardDownload

Elevator group supervisory control system using genetic network programming - Ranking processing and node function optimization

by Sandor Markon

2025

Genetic Network Programming (GNP) has been proposed as a new method of evolutionary computations and its basic characteristics are studied. Recently, GNP is also applied to Elevator Group Supervisory Control System (EGSCS) which is a... more

descriptionView Paper arrow_downwardDownload

An Efficient Model for Network Intrusion Detection System based on an Evolutionary Computational Intelligence Approach

by Ruba Soundar K

2024

Intrusion Detection systems are increasingly a key part of system defence. Various approaches to Intrusion Detection are currently being used but false alarm rate is higher in those approaches. Network Intrusion Detection involves... more

descriptionView Paper arrow_downwardDownload

Herald NAMSCA 1, 2018

by Irshad Ullah

2024

Computing science or computer Science is the study of theoretical foundation of knowledge and computation, with practical methods for their implementation and execution in computer environment. It may be described as the meaningful study... more

Other method know is sampling minimize the scanning of database to one in their best case while two in their worst case.'° In this technique a portion of the database brought into the main memory first and then it is used to generate the large itemsets by using well known algorithms like apriori. All these approaches are based on the idea that data is stored in the database in some structure like flat-file or may be TID while the actual data in the database is not in that structure. All these algorithms and approaches work on the idea that first to find the large itemsets. In all these approaches Apriori is seem to be the most efficient algorithms. In this research The focus will be on Apriori algorithm. Here two algorithms were applied Apriori and the algorithm produced by us that is SI algorithm. "'

Variables of the type binary can take only two values true or false either 0 or 1. If it is think that al variables having the same weight then a 2 by 2 contingency table is given Table3: Contingency table for binary variables Logical data analysis was firstly developed in the late eighties. Its aim is to discover hidden knowledge pattern in the database of Boolean type. The aim of the LAD is to analyze the Boolean data but then it is extended for numerical data by the introduction of Binarization. The method of LAD was extended to numerical data by the process of Binarization in which numerical data is transferred to the binary values like zero and ones. 16

More amount of data were loaded which raise number of records into (2000) these 2000 were given to apriori algorithm. Its produce the same largest set of item as was produced with 1000 records given below.

Now after the implementation of the database with the two algorithms, it is much clear that the results are accurate and efficient. Changing size of the database or of the algorithm does not effect the accuracy although the algorithm uses different strategies for the production purpose. “Now as the performance were evaluated by changing the size of the database. Next to make the process more reliable the input threshold has to be changed. For that purpose for SI algorithm the input value was changed from .80 to .70 and from twenty to thirty for apriori algorithm. Now after changing the algorithm were applied to scan the database to produce the largest item list. Thus, the item set produced (large) by both of the algorithms are: T1 72 JT4

Now the SI algorithm was implemented on the same amount of data

descriptionView Paper arrow_downwardDownload

An Assessment of Time Performance Metrics of Lift Systems in Selected Public High-Rise Buildings in Abuja, Nigeria

by Olugbenga Ayeni

2023

Lift systems in high-rise buildings are often confronted with problems associated with optimum service delivery. The optimum performance of these systems in public high-rise buildings enhances the efficiency of vertical transportation... more

descriptionView Paper arrow_downwardDownload

Elevator group supervisory control system using GNP with reinforcement learning based on normalized information

by Sandor Markon

2023

Genetic Network Programming (GNP) has been proposed as a new method of evolutionary computations and its basic characteristics are studied. Recently, GNP is also applied to Elevator Group Supervisory Control System (EGSCS) which is a... more

descriptionView Paper arrow_downwardDownload

Elevator group supervisory control system using genetic network programming - Ranking processing and node function optimization

by Sandor Markon

2023

Genetic Network Programming (GNP) has been proposed as a new method of evolutionary computations and its basic characteristics are studied. Recently, GNP is also applied to Elevator Group Supervisory Control System (EGSCS) which is a... more

descriptionView Paper arrow_downwardDownload

Elevator Group Supervisory Control System Using Genetic Network Programming with Macro Nodes and Reinforcement Learning

by Sandor Markon

2023, The transactions of the Institute of Electrical Engineers of Japan.C

Elevator Group Supervisory Control System (EGSCS) is a very large scale stochastic dynamic optimization problem. Due to its vast state space, significant uncertainty and numerous resource constraints such as finite car capacities and... more

Fig.1. Basic Structure of an EGSCS using GNP with RL Keywords: elevator group supervisory control system, genetic network programming, reinforcement learning, impor- tance weight

Fig. 2. Flowchart of evolution and learning

Fig.4. Basic Structure of an EGSCS using GNP with RL

Fig.5. Fitness Curves in each Traffic Flow one kind of judgment node and two kinds of processing nodes, with 10 nodes of each kind. Moreover, the node size is also fixed for simplicity.

Table 4. Performance Comparison of Different Methods in Simulations

Table 3. Parameters for GNP with RL 4. Simulation Results

Table 1. Specifications of EGSCS Simulator Table 2. Ratio of Passengers in Each Traffic Flow

descriptionView Paper arrow_downwardDownload

Double-deck elevator systems using genetic network programming based on variance information

by Sandor Markon

2023, SICE Annual Conference 2007

In order to increase the transportation capability of elevator group systems in high-rise buildings without adding elevator installation space, double-deck elevator system (DDES) is developed as one of the next generation elevator group... more

descriptionView Paper arrow_downwardDownload

Comparing Decentralized Ongoing Task Allocation

by Gita Sukthankar

2023, 2020 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT)

Real-world domains can often benefit from adaptable decentralized task allocation through emergent specialization of large groups of agents. Agents working within a shared environment can use environmental cues as a means of indirect... more

descriptionView Paper arrow_downwardDownload

Comparing Decentralized Ongoing Task Allocation

by Vera Kazakova

2023, 2020 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT)

Real-world domains can often benefit from adaptable decentralized task allocation through emergent specialization of large groups of agents. Agents working within a shared environment can use environmental cues as a means of indirect... more

descriptionView Paper arrow_downwardDownload

Comparing Decentralized Ongoing Task Allocation

by Gita Sukthankar

2023, 2020 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT)

Real-world domains can often benefit from adaptable decentralized task allocation through emergent specialization of large groups of agents. Agents working within a shared environment can use environmental cues as a means of indirect... more

descriptionView Paper arrow_downwardDownload

Genetic Network Programming with Reinforcement Learning Using Sarsa Algorithm

by Thu Thu

2023, 2006 IEEE International Conference on Evolutionary Computation

A new graph-based evolutionary algorithm called Genetic Network Programming, GNP has been proposed. The solutions of GNP are represented as graph structures, which can improve the expression ability and performance. In addition, GNP with... more

descriptionView Paper arrow_downwardDownload

Improving coordination via emergent communication in cooperative multiagent systems: A Genetic Network Programming approach

by Mehran Ghaziasgar

2022, 2009 IEEE International Conference on Systems, Man and Cybernetics

The design and development of strategies to coordinate the actions of multiple agents is a central research issue in the field of multiagent systems. To address this issue, in our previously reported research we proposed a novel... more

descriptionView Paper arrow_downwardDownload

Trends in the Development of Machinery and Associated Technology

by Berna Bolat

2022

Screw conveyors are used for the transmission of powder, small-grained and liquid materials in the different areas of industry. Screw conveyors have rotating helical shafts inside fixed a through or pipe. The goods are delivered with the... more

descriptionView Paper arrow_downwardDownload

Double-Deck Elevator System Uing Genetic Network Programming with Genetic Operators Based on Pheromone Information

by Sandor Markon

2022, 2008 Eighth International Conference on Hybrid Intelligent Systems

Genetic Network Programming (GNP), one of the extended evolutionary algorithms was proposed, whose gene is constructed by the directed graph. GNP is distinguished from other evolutionary techniques in terms of its compact structure and... more

descriptionView Paper arrow_downwardDownload

Double-deck elevator systems using Genetic Network Programming with reinforcement learning

by Sandor Markon

2022

In order to increase the transportation capability of elevator group systems in high-rise buildings without adding elevator installation space, double-deck elevator system (DDES) is developed as one of the next generation elevator group... more

descriptionView Paper arrow_downwardDownload

Discovering numeric association rules via evolutionary algorithm

by Jacinto Mata

2022

Association rules are one of the most used tools to discover relationships among attributes in a database. Nowadays, there are many efficient techniques to obtain these rules, although most of them require that the values of the... more

descriptionView Paper arrow_downwardDownload

Discovery of association rules from medical data-classical and evolutionary approaches

by Halina Kwasnicka

2022, Conference Proceedings pp

Abstract. The paper presents a method of association rules discovering from medical data using evolutionary approach. The elaborated method (EGAR) uses a genetic algorithm as a tool of knowledge discovering from a set of data, in the form... more

descriptionView Paper arrow_downwardDownload

Association Rules Algorithms for Data Mining Process Based on Multi Agent System

by Mohammed Talibi Alaoui

2022, Machine Learning for Networking

In this paper, we present a collaborative multi-agent based system for data mining. We have used two data mining model functions, clustering of variables in order to build homogeneous groups of attributes, association rules inside each of... more

descriptionView Paper arrow_downwardDownload

An adapted covering algorithm approach for modeling airplanes landing gravities

by Chaochang Chiu

2021, Expert Systems with Applications

The covering algorithm is an often-used rule induction method. The main shortcoming of this algorithm is its inability to handle continuous-type data. The present research proposes a novel method that integrates genetic algorithms with... more

descriptionView Paper arrow_downwardDownload

Large-Scale Association Rule Discovery from Heterogeneous Databases with Missing Values using Genetic Network Programming

by Eloy Gonzales

2021

Association Rule Mining is an important data mining task and it has been studied from different perspectives. Recently multi-relational rule mining algorithms have been developed due to many real-world applications. However, current work... more

Figure 3. A connection of nodes in GNP for association rule mining with missing values

Figure 4. Basic structure of GNP for association rule mining

Figure 6. Number of extracted rules (generations=100) Figure 5. Number of extracted rules (min confidence=0.8)

Figure 7. Processing Time (min confidence=0.8)

Figure 8. Number of Rules with different min support and min confidence

Figure 9. Processing Time with different min support and min confidence

INFORMATION OF THE ORIGINAL DATASETS Table III

NT; : node type (Start node=0; Judgment node=1; Processing node=2) ID; : identification number; d, dj: delay time; Cj: connected node

EXAMPLE OF DATABASE WITH MISSING DATA Table I is an example of an incomplete database which contains missing values. A; is an attribute in the database. Missing data is represented as ’m”’, a different value of 1 or 0. Most of the conventional association rule mining al- gorithms assume that databases are complete. Generally, databases are pre-processed in order to eliminate missing values or to replace them with an average or other statistical measures because the main problem in such kind of datasets is the difficulty for calculation of measures such as support, confidence and cosine.

descriptionView Paper arrow_downwardDownload

Association Rule Mining from Large and Heterogeneous Databases with Uncertain Data using Genetic Network Programming

by Eloy Gonzales

2021

Association Rule Mining is one of the most important tasks in data mining and it has been deeply studied during last years. Recently several rule mining algorithms have been developed due to many real-world applications. Most of these... more

descriptionView Paper arrow_downwardDownload

Mining Fuzzy Association Rules: A General Model Based on Genetic Network Programming and its Applications

by Eloy Gonzales

2021, IEEJ Transactions on Electrical and Electronic Engineering

The initiative of combining association rule mining with fuzzy set theory has been applied frequently in recent years [1-5]. The original idea comes from dealing with quantitative attributes in a database, where discretization of the... more

Fig. 1. The basic structure of GNP individual 3.1. Chromosome representation The basic structure of GNP is shown in Fig. 1. GNP is composed of three types of nodes that are connected on a network structure: a start node, judgment nodes (diamonds) and processing nodes (circles). Judg- ment nodes are the set of Jj, J2,..., J), which work as if-then conditional decision functions and they return judgment results for assigned inputs and determine the next node to be executed. Processing nodes are the set of P,, P,..., P,, which work as action/processing functions. The start node determines the first node to be executed. The node transition begins from the start node; however there are no terminal nodes. After the start node is executed, the next node is determined according to the node’s connections and judgment results.

NT, : node type (0: start node; 1: processing node; 2: judgment node) ID, : identification number; d; : delay time; Cy: connected node

From two parents, the selected nodes and its connections are swapped each other by crossover rate P...

The characteristics of GNP are described as follows.

Fig.5. Membership functions for attribute 7 Fig.6. Fuzzy attributes obtained from quantitative databases

ol. PUZZY MCMVCrsnip LUNCUONS LOF Nhanaduns quan- titative attributes At first, the values of all quantitative attributes of the database are fuzzified, for example, into three linguistic terms (e.g., low, middle and high). These linguistic terms are defined by the combination of trapezoidal and triangu- lar membership functions symmetrically spaced (see Fig. 5). Each quantitative attribute is associated with its own membership func- tions (see Fig. 6). In addition, the parameters of the membership

Fig.7. Example of the mutation of the fuzzy parameters

Fig.8. Basic GNP structure for fuzzy association rule mining

Fig.9. Example of GNP individual for extract fuzzy rules

Fig. 10. Transition from one judgment node to another using fuzzy values using probability P;

Fig. 11. Updating the pool of fuzzy rules in every generation 4.6. Updating the pool of fuzzy rules The mined fuzzy class association rules are stored in a pool through gen- erations as Fig. 11 shows. When an important rule is extracted by GNP, it is stored in a pool of rules with its support, confidence, x7 values, and fuzzy parameters. Occasionally, a fuzzy rule already stored in the pool can be once again extracted, and in that situa- tion, the membership functions and x? values might be changed.

Fig. 12. Framework for searching fuzzy membership functions and fuzzy class association rules

Figure 13 shows the total number of fuzzy rules extracted when the maximum number of generations is set to 100, while Fig. 14

Fig. 15. Number of extracted fuzzy rules in CRX DB for dif- ferent minimum support values Fig. 14. Number of extracted fuzzy rules in each database

Fig. 16. Classification accuracy for CRX DB for different min- imum support values

Fig. 17. Processing time for CRX DB for different minimum support values

Table I. Support and confidence of fuzzy rules Table II. The contingency of X and Y

The values in bold face mean that the classification accuracy of the proposed method is highest compared with the other methods. Table V. Comparison of the classification accuracy (%)

descriptionView Paper arrow_downwardDownload

Association Rule Mining for Continuous Attributes using Genetic Network Programming

by Eloy Gonzales

2021, IEEJ Transactions on Electrical and Electronic Engineering

Most of the existing association rule mining algorithms are able to extract knowledge from databases with attributes of binary values. However, in real-world applications, databases are usually composed of continuous values such as... more

descriptionView Paper arrow_downwardDownload

Association Rule Mining for Continuous Attributes using Genetic Network Programming

by Eloy Gonzales

2021, Ieej Transactions on Electrical and Electronic Engineering

Most of the existing association rule mining algorithms are able to extract knowledge from databases with attributes of binary values. However, in real-world applications, databases are usually composed of continuous values such as... more

descriptionView Paper arrow_downwardDownload

A study on techniques for the automatic generation of membership functions for pattern recognition

by Carolina Monard

2021

The task of automatically generating suitable membership functions from data is fundamental for the appropriate performance of fuzzy systems that use a fuzzy rule base. In this paper we provide an overview of some methods found in the... more

Figure 1. Merging two membership functions [Manley-Cooke and Razaz 2007]. set (support), and i1) a value for a symbolic translation, which enables fuzzy sets to be am- plified to its laterals by a displacement. The symbolic translation of a linguistic term is a number within the interval |—0.5, 0.5) that expresses the domain of a linguistic term when it is moving between its two lateral linguistic terms. Figure 2-a shows the symbolic trans- lation of a linguistic term represented by the pair ($2, —0.3), considering a set S with five linguistic terms represented by their ordinal values ({0,1,2,3,4}). Figure 2-b presents the representation of the original fuzzy term S52, represented by a triangular shaped set in black, and another interpretation of this same set according to the pair ($2, —0.3), which defines a displacement of —0.3 to the left (shown in light blue).

Figure 2. Symbolic translation of a linguistic term and lateral displacement of the involved membership function [Alcala et al. 2007].

descriptionView Paper arrow_downwardDownload

Genetic Network Programming with Reinforcement Learning and Its Application to Making Mobile Robot Behavior

by Thu Thu

2021, IEEJ Transactions on Electronics, Information and Systems

A new graph-based evolutionary algorithm named "Genetic Network Programming, GNP" has been proposed. GNP represents its solutions as graph structures, which can improve the expression ability and performance. Since GA, GP and GNP already... more

descriptionView Paper arrow_downwardDownload

Discovering numeric association rules via evolutionary algorithm

by Cristobal Riquelme

2021

Association rules are one of the most used tools to discover relationships among attributes in a database. Nowadays, there are many efficient techniques to obtain these rules, although most of them require that the values of the... more

descriptionView Paper arrow_downwardDownload

Genetic network programming with reinforcement learning and its application to making mobile robot behavior

by Thu Thu

2021, 電気学会論文誌 C (電子・ …

A new graph-based evolutionary algorithm called Genetic Network Programming, GNP" has been proposed. The solutions of GNP are represented as graph structures, which can improve the expression ability and performance. In addition,... more

descriptionView Paper arrow_downwardDownload

Exploration of genetic network programming with two-stage reinforcement learning for mobile robot

by TELKOMNIKA JOURNAL

2020, TELKOMNIKA Telecommunication Computing Electronics and Control

This paper observes the exploration of Genetic Network Programming Two-Stage Reinforcement Learning for mobile robot navigation. The proposed method aims to observe its exploration when inexperienced environments used in the... more

descriptionView Paper arrow_downwardDownload

Operator Self-adaptation in Genetic Programming

by Robert Mckay

2016, Lecture Notes in Computer Science

We investigate the application of adaptive operator selection rates to Genetic Programming. Results confirm those from other areas of evolutionary algorithms: adaptive rate selection out-performs non-adaptive methods, and among adaptive... more

descriptionView Paper arrow_downwardDownload

Group Elevator Scheduling With Advance Information for Normal and Emergency Modes

by Peter Luh and

2016, IEEE Transactions on Automation Science and Engineering

Group elevator scheduling has long been recognized as an important problem for building transportation efficiency, since unsatisfactory elevator service is one of the major complaints of building tenants. It now has a new significance... more

Fig. 1. Rolling horizon scheme with a look-ahead window T.

Fig. 2. A model of single elevator dynamics.

Fig. 4. Definitions of time metrics. For individual elevators, capacity limits are formulated a: linear inequality constraints

Fig. 5. The decomposition and coordination approach. elevator subproblems, (4) is transformed into an additive form in terms of elevators by using (1)

Fig. 6. Illustration of the local search. By collecting all the terms related to elevator 7 from (6), the subproblem 7 is obtained as

Fig. 7. Illustration of the three-passage heuristics.

Fig. 8. Stagewise costs in stage k for different cases.

With (23), the problem is converted to a separable form, anc our decomposition and coordination approach can be applied. A Lagrangian function is first obtained by relaxing the assignmen' constraints (1) with nonnegative multipliers {;}, and the nev egress time constraints (23) with nonnegative multipliers {j.; }

RESULTS FOR LOW INTENSITY INTERFLOOR TRAFFIC Fig. 9. Passenger arrival rates for each interval.

RESULTS FOR VARIOUS TRAFFIC PATTERNS WITH VARYING ARRIVAL RATES Fig. 10. Average service times for up-peak with different window lengths. TABLE IX

RESULTS FROM DIFFERENT METHODS FOR SUBPROBLEM SOLVING Furthermore, for each time window, interfloor has the smallest average service time among the three traffic patterns, followed by down-peak. This is reasonable since for interfloor traffic, an elevator can serve passengers on its way up and down, and for down-peak traffic on its way down, but for up-peak traffic only at the lobby.

RESULTS FOR VARIOUS TRAFFIC PATTERNS WITH DIFFERENT WINDOW LENGTHS

RESULTS FROM BATCH OPTIMIZATION AND 1D INCREMENTAL OPTIMIZATION TABLE VI RESULTS FOR UP-PEAK WITH DIFFERENT WINDOW LENGTHS

rescheduling points over a set of passengers. For incremental optimization, however, the assignment decision is optimized upon the arrival of each passenger assuming that no future traffic information is available. To compare these two opti- mization schemes, a window length of zero is used with a rescheduling interval of 7.5 s, and results are summarized in Table VI. With incremental optimization, the average service time is 61.8 s with a standard deviation of 0.83. Both metrics are higher than those obtained by using batch optimization, al- though the average CPU time of 2.2 s is less than 5.7 s obtained before. This indicates that incremental optimization cannot take the advantage of optimizing over a set of passengers.

RESULTS FROM THE UP-PEAK DISPATCHING POLICY AND OUR METHOD TABLE X

RESULTS FOR EVACUATION WITH ELEVATORS AND WITHOUT ELEVATORS TABLE XI

descriptionView Paper arrow_downwardDownload

Exact calculation of expected waiting times for group elevator control

by Daniel Nikovski

2015

We present an efficient algorithm for exact calculation and minimization of expected waiting times of all passengers using a bank of elevators. The dynamics of the system are represented by a discrete-state Markov chain embedded in the... more

to the last possible location at which a car should start decelerating if it is to stop at a particular floor in its direction of motion. A particular path of a car dur- ing its round-trip always consists of a finite number of such segments, whose endpoints are branching or resting points. Consequently, if the waiting time on each such segment can be computed, it can be reused for the computation along many paths that include that segment. Reusing the costs on all individual segments can be achieved by embedding a discrete Markov chain into the original system of elevator movement, which in it- self operates in continuous time and space. A Markov chain consists formally of a finite number of states S;,7 € [1,.N,], an immediate cost C;,; of the transi- tion between each pair of states S; and S, a matrix P,; of the probabilities of transition between states S; and S;, and a distribution 7(S;) which specifies the probability that the system would start in state S; [3]. Te cetieawe Caw tha Alveecew 8 he AAnwleeesew 6S wha TA

descriptionView Paper arrow_downwardDownload

Discretization Algorithm that Uses Class-Attribute Interdependence Maximization

by Lukasz Kurgan and

2015, International Conference on Artificial Intelligence

Most of the existing machine learning algorithms are able to extract knowledge from databases that store discrete attributes (features). If the attributes are continuous, the algorithms can be integrated with a discretization algorithm... more

Table 1. 2-D frequency matrix for attribute F and discretization scheme D discretization variable D for attribute F is defined as:

Table 2. Major properties of datasets considered in the experimentation

Table 3. Comparison of the six discretization schemes using four continuous and mixed-mode datasets (bolded values show the best results) Table 3 shows the results of discretizing the datasets using all considered discretization schemes. The CAIM algorithm achieved the highest class-attribute interdependency for 3 out of 4 datasets, and for wav datasets was the second highest. That verifies that the greedy approach and the CAIM criterion work in practice, and results in higher interdependence between class and attribute variables than the interdependence achieved bv other alaorithms. 3.1. Analysis of the results

Table 4. Comparison of the accuracies achieved by the CLIP4 algorithm for the four datasets using the six discretization schemes (bolded are the best results, hea and pid results are averaged over 10CV)

descriptionView Paper arrow_downwardDownload

Effects of passenger's arrival distribution to double-deck elevator group supervisory control systems using genetic network programming

by Sandor Markon

2015, Proceedings of the 9th annual conference on Genetic and evolutionary computation - GECCO '07

The Elevator Group Supervisory Control Systems (EGSCS) are the control systems that systematically manage three or more elevators in order to efficiently transport the passengers in buildings. Double-deck elevators, where two cages are... more

descriptionView Paper arrow_downwardDownload

A Double-Deck Elevator Group Supervisory Control System with Destination Floor Guidance System Using Genetic Network Programming

by Sandor Markon

2015, IEEJ Transactions on Electronics, Information and Systems

descriptionView Paper arrow_downwardDownload

Validation and Optimization of an Elevator Simulation Model with Modern Search Heuristics

by Sandor Markon

2015, Operations Research/Computer Science Interfaces Series

Elevator supervisory group control (ESGC) is a complex combinatorial optimization task that can be solved by modern search heuristics. To reduce its complexity and to enable a theoretical analysis, a simplified ESGC model (S-ring) is... more

Figure 5.1. The S-ring as an elevator system. The sites are numbered from 0 to n—1. Metaheuristics: Progress as Real Problem Solvers

Figure 5.2. Event graph. The thin arrow represents an event at the beginning (ar- rival) scheduling an event at the end of the arrow (departure) immediately. The heavy arrows indicate that the event at the beginning of the arrow may schedule the event at the end of the arrow.

Figure 5.3. Function depart. The probability of an arrival event is set to 0.5.

Figure 5.4. Schematic representation of algorithm based validation. A regression model describes the functional relationship between the parameter setting of an al- gorithm and its expected performance. Two optimization problems are equivalent, if there are no interactions between the problem and model parameters.

Figure 5.5. Two NN controllers dealing with different update methods (sequential, random, quasi-parallel) and arrival rates on an S-ring with 20 floors and 6 servers. Left hand: Default controller, all weights are 1.0, right hand: ES-optimized controller. Each point represents a simulation run with 1 million steps. Lower values are better.

Figure 5.6. Default (left) and optimized NN controller performance on a wide range of arrival rates and server capacities on an S-ring with 20 floors and 6 servers. Each point represents a simulation run with 1 million steps. Lower values are better.

Table 5.1. The triple in the first column represents the state of the actual site: Cus- tomer waiting, server present, and server present on the next floor. The probability of a state change to the state in the fourth column is given in the second column Columns three and five denote the decision and the reward respectively. I.e., the server has to make a decision (to ‘take’ or to ‘pass’ the customer) if there is a cus- tomer waiting (1xx), and if there is a server present on the same floor (11x) but nc server on the next floor (110). the unobservable case. Next we will introduce some elementary policies: — The most obvious heuristic policy is the greedy one: When given the choice, always serve the customer: 79(0) = 1. Rather counter-intuitively, this policy is not optimal, except in the heavy traffic (p > 0.5) case. This means that a good policy must bypass some customers occasionally. The greedy-policy does not take any information about the state of the sys- tem into account.

descriptionView Paper arrow_downwardDownload

Double-deck Elevator Group Supervisory Control System using Genetic Network Programming with Ant Colony Optimization

by Sandor Markon

2015, 2007 IEEE Congress on Evolutionary Computation

descriptionView Paper arrow_downwardDownload

Double-deck elevator systems using genetic network programming based on variance information

by Sandor Markon

2015, SICE Annual Conference 2007

In order to increase the transportation capability of elevator group systems in high-rise buildings without adding elevator installation space, double-deck elevator system (DDES) is developed as one of the next generation elevator group... more

descriptionView Paper arrow_downwardDownload

A Double-Deck Elevator Group Supervisory Control System Using Genetic Network Programming

by Sandor Markon

2015, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews)

descriptionView Paper arrow_downwardDownload

Double-deck elevator system using genetic network programming with genetic operators based on pheromone information

by Sandor Markon

2015

Genetic Network Programming (GNP), one of the extended evolutionary algorithms was proposed, whose gene is constructed by the directed graph. GNP is distinguished from other evolutionary techniques in terms of its compact structure and... more

descriptionView Paper arrow_downwardDownload

Double-deck elevator systems using Genetic Network Programming with reinforcement learning

by Sandor Markon

2015

In order to increase the transportation capability of elevator group systems in high-rise buildings without adding elevator installation space, double-deck elevator system (DDES) is developed as one of the next generation elevator group... more

descriptionView Paper arrow_downwardDownload

Elevator group supervisory control systems using genetic network programming

by Sandor Markon

2015, … . CEC2004. Congress on

Genetic network programming (GNP) has been proposed as a new method of evolutionary computation. Until now, GNP has been applied to various problems and its effectiveness was clarified. However, these problems were virtual models, so the... more

descriptionView Paper arrow_downwardDownload

Validation and optimization of an elevator simulation model with modern search heuristics

by Mike Preuss

2015, Metaheuristics: progress as real problem solvers

Elevator supervisory group control (ESGC) is a complex combinatorial optimization task that can be solved by modern search heuristics. To reduce its complexity and to enable a theoretical analysis, a simplified ESGC model (S-ring) is... more

descriptionView Paper arrow_downwardDownload

Design of modern elevator group control systems

by Tsung-Che Chiang

2015, Proceedings 2002 IEEE International Conference on Robotics and Automation (Cat. No.02CH37292)

The advantages of our EGCS are shown through extensive simulation results. We make a comparison between our EGCS and the previous one, which shows that our results are quite satisfactory and superiol:

Fig. 1 The overview of the elevator system.

rescheduling is necessary. Here we propose a ptiority-based mechanism which reschedules hall calls without evaluating all possible assignments. This mechanism arranges the order of hall calls to do HCAM.

is helpful to relieve other elevators’ load. If other elevators do not need its participation, we send this elevator to the floor where passenger arrival rate is high.

In this paper, we proposed a new viewpoint to design the modern EGCS. Thinking from passengers’ viewpoint provides much more information to make good schedules for the elevators. This is achieved by construction of temporal traffic database, passenger-based evaluation, and the utilization of the capacity constraints.

Table 1: Major differences between previous and current EGCS Table 2: Comparison of performance during four traffic patterns

descriptionView Paper arrow_downwardDownload

Online Rule Updating System Using Evolutionary Computation for Managing Distributed Database

by Wirarama Wedashwara

2015

This research proposes a decision support system of database cluster optimization using genetic network programming (GNP) with on-line rule based clustering. GNP optimizes cluster quality by reanalyzing weak points of each cluster and... more

descriptionView Paper arrow_downwardDownload

Combination of Genetic Network Programming and Knapsack Problem to Support Record Clustering on Distributed Databases

by Wirarama Wedashwara

2015

This research involves implementation of genetic network programming (GNP) and knapsack problem (KP) as a decision support system for record clustering in distributed databases. The objective is to distribute big data to certain sites... more

descriptionView Paper arrow_downwardDownload

Implementation of Genetic Network Programming and Knapsack Problem for Record Clustering on Distributed Database

by Wirarama Wedashwara

2015

This research involves implementation of genetic network programming (GNP) and knapsack problem (KP) to solve record clustering on distributed databases. The objective is to distribute big data to certain sites with the limited amount of... more

descriptionView Paper arrow_downwardDownload

Operator Self-adaptation in Genetic Programming

by MinHyeok Kang

2014

We investigate the application of adaptive operator selection rates to Genetic Programming. Results confirm those from other areas of evolutionary algorithms: adaptive rate selection out-performs non-adaptive methods, and among adaptive... more

descriptionView Paper arrow_downwardDownload

Genetic Network Programming

Related Topics