Volume 17-Issue4-2008-SAAD

Sawsan SAAD, Hayfa ZGAYA
LAGIS UMR 8146, Ecole Central de Lille, France

Slim HAMMADI
LAGIS UMR 9146, Ecole Central de Lille, France

Abstract: In this paper, we address the problem of negotiation process in a multi-agents system by using ontologies. Therefore, we present an ontology solution based on the knowledge management system for semantic heterogeneity. The proposed solution prevents the misunderstanding during the negotiation process through the agents’ communications. Our approach aims to enable agents able to understand each other when using these ontologies. Thus, we propose a general architecture for Negotiation process which uses Ontology-based Knowledge Management System (NOKMS). This architecture consists of three layers: the Negotiation Layer (NL) that describes the negotiation process between the Initiator Static Agents (ISAs) and the Participant Mobile Agents (PMAs) by using suitable ontologies, the Semantic Layer (SEL) contains the semantic translator which uses in the case of misunderstanding of the sent messages between the agents, and the last one is the Knowledge Management Systems Layer (KMSL) which bases on the Intelligent Knowledge Base (IKB) to give the flexibility to our negotiation ontology. In addition, we will illustrate an agent architecture which helps our architecture on applying the different operations in the different layers. Finally, we present a case study which applies our architecture on the Multimodal Transport Information System (MTIS) project where we will show two scenarios applicable: the first uses our negotiation ontology architecture in one transport system, and the second applies this architecture on the multi-transport systems. These case studies show that the proposed NOKMS improves the execution of negotiation process in multi-agents systems in order to satisfy the transport customers.

Keywords: Multi-Agents Systems, Negotiation, Ontology, Knowledge Management System, Transport Information System.

Sawsan Saad is currently a doctorate in LAGIS laboratory at the high France School, Ecole Centrale de Lille (EC-Lille) in the Optimization, the Artificial Intelligence and the Logistic field. Born in Salamieh (Syria) in 1979. She received the BSc degree in Computer Engineering from Al-Baath University (Syria) in 2003. She obtained her MSc degree in Information Systems in 2007 from INSA de Lyon (France). She was member of the Organizing Committees of international congress on “Logistic and Transport” LT’2007.

Dr. Hayfa Zgaya is teacher-researcher in Computing Science in LAGIS laboratory within the high France School Ecole Centrale de Lille (EC-Lille) where she obtained her PhD degree in July 2007. Her main research areas are the Optimization, the Artificial Intelligence and the Logistic field. Born in Tunis in 1978, she obtained her master’s degree in computer science in November 2001 from the France polytechnic high school of Nantes University. She takes an active part in the national working group ORT of GDR MACS and she was responsible of the Organizing Committees of international workshops MHOSI’ 2005, LT’2006 and LT’2007.

Slim Hammadi is a full Professor of production planning and control at the Ecole Centrale de Lille (French “Grande Ecole”). Born in Gafsa (Tunisia) in 1962, he has obtained by 1988 the Master degree in Computer science from the University of Lille (France). Pr Hammadi obtained a P.h.D degree in job-shop scheduling and control in 1991 at Ecole Centrale de Lille. He is a senior member of IEEE/SMC and has served as a referee for numerous journals including the IEEE Transactions on SMC. Pr. S. Hammadi was Co-Organizer of a Symposium (IMS) of the IMACS/IEEE SMC Multi conference CESA’98 held in Hammamet (Tunisia) in April 1998. He has organized several invited sessions in different SMC conferences where he was session chairman. He was chairman of the International congress on “Logistic and Transport” LT’2004, MHOSI’2005, LT’2006 and LT’2007. His teaching and research interests focus on the areas of production control, production planning, computer science, discrete and dynamic programming and computer integrated manufacturing.

>>Full text
CITE THIS PAPER AS:
Sawsan SAAD, Hayfa ZGAYA, Slim HAMMADI, Novel Ontology Model for Communicating Heterogeneous Negotiation Mobile-Agent in a Transport Environment, Studies in Informatics and Control, ISSN 1220-1766, vol. 17 (4), pp. 333-352, 2008.

1. Introduction

This work belongs to the French national project VIATIC.MOBILITE from the industrial cluster I-TRANS “France highlights leading-edge technology in rail systems and innovative transport” . In fact; it has become much more important for public transportation companies to provide an adequate Level Of Service to their customers. This is due to the growing competition in the public transport market and also to the privatization of the transport companies. The great difficulty related to the traffic management in such systems is allied to the respect of the planned departure and arrival times of the vehicles at the different stops of the network. In fact, many incidents can occur and force a customer to wait longer, which decreases the level of service. Hence, operational decisions have to be taken in real-time by human decision makers, called regulators. However, these operators are overloaded with information that they have to treat immediately in order to find the relevant decisions that result in new vehicle schedules. Decision Support Systems (DSS) are computer technology solutions that can be used to support complex decision making and problem solving. Decision making is the study of how decisions are actually taken and how they can be better or more successfully taken. In order to realize the necessity of a DSS for a Transportation System, it is important to grasp the problems related to a real-time management of the traffic [20].

In our Multimodal Transport Information Syste (MTIS) [1] project, we presented a Multi-Agent Decision Support Systems that provides the regulators with the relevant decisions to undertake in case of disturbances. In fact, scheduling can be defined as a problem of finding the optimal sequence for executing a finite set of operations under a certain set of constraints which must be satisfied. A scheduler usually attempts to maximize the utilization of individuals and / or machinery and minimize the time required to complete the entire process being scheduled. Therefore the scheduling problem is very hard to solve [27]. A Genetic Algorithms (GAs) have been used to solve this problems in our MTIS. The proposed multi-agent system is based on metaheuristics for the research and the composition of the services; services research is based on the Mobile Agent paradigm (MA) using this dynamic optimization algorithm for the MA Workplans design. The first step of optimization prepares the MA routes, taking into account the network state. The services composition uses evolutionary algorithms to optimize the responses in terms of costs and delays, knowing that a response to a user request must respect a fixed due date with a reasonable cost. We also designed and optimized the management of the data flow of the users’ requests, which can be simultaneous and numerous. We developed also a negotiation protocol intended for the transport area which permits the agents to negotiate when perturbations may exist and as a result the system needs to reassign news nodes. The negotiation protocol uses messages to exchange the information. Those messages are exchanged between the Scheduler Agents (SAs), representing the initiators of the negotiation, and the Intelligent Collector Agents (ICAs), representing the participants of the negotiation. This protocol has studied before only the cases of the simple messages and it proposed ontology without illustrating it, and this later didn’t include the solutions when the agents ICAs did not understand the messages sent from the SA agent In this paper, we propose an approach that aims to improve the protocol of the negotiation of the multi-agents systems which has been proposed in the previous work [2]; we present an ontology solution based on the knowledge management system for semantic heterogeneity. The proposed solution prevents the misunderstanding during the negotiation through the agents’ communications. Our approach aims to make the agents able to understand each other when using these ontologies.