Tunnels in an ontological perspective

0 downloads 0 Views 644KB Size Report
The Application Of Artificial Intelligence To Civil And Structural Engineer- ing. Civil-Comp Press .... The official Manual of Road Tunnels Traffic and Fire Safety of the Nor- wegian Traffic Society .... book of Metaphysics and Ontology. Philosophia ...
Tunnels in an ontological perspective M. Cristani

C.E. Majorana, V. Salomoni

Dipartimento di Informatica,

Dipartimento di Costruzioni e Trasporti,

Universit` a di Verona

Universit` a di Padova

strada Le Grazie 15, Verona

via Marzolo, 9 Padova

[email protected]

{majorana,salomoni}@caronte.dic.unipd.it

In this paper we shall illustrates one of the core parts of the ontological layer of an application, which is under development [1], whose purpose is to support upgrading of long tunnels. This application will be an ontology-driven web decision support system. The reasons for supporting this process in such a semi-automatic way are many: • The aspects that have to be considered when applying an upgrade procedure are so many that in practice it is impossible to manage them without an automated support; • The number of involved experts is high, and they usually do not speak the same technical language, and often do not speak the same natural language as well, therefore we need a support for a a unified model of knowledge, that will necessarily be a web-based one, due to the geographic spread of experts; • It is not possible to share data of di↵erent experts without a direct support of unified processing service. Many of the processing utilities presently employed or under development have proprietary input and output data formats, forcing therefore the experts to transfer data to each other either through complex semi-automated translation processes or by means of human understanding, often supported by printed data sheets. In this paper we illustrate the conceptual organisation of data describing tunnels. This describes a tunnel as an object in a model of knowledge developed in OWL, deployed by means of an Ontology Editor. We used Protege-2000, that is an open source system. The top-level of this ontological model is described in Figure 1. The aspects included in this description are fundamentally five: the functional and prescriptive requirements of tunnels, the tunnel safety features, the tunnel structure and geometry, the data description of Mass-Heat Flow analysis, the notion of documents to be deployed in the IntelliTun application. 1

Figure 1: The top level of the ontology of tunnels implemented in OntoTunnel. For all the above mentioned aspects, the ontological perspective upon tunnels we have undertaken, is essentially based on the needs of the application, and in the sense which is proposed in the current literature about formal ontology is a task-oriented ontology. This justifies some technicalities of the structure which are not needed in a more general and less application-oriented specification.

References [1] Matteo Cristani, Gabriel A. Khoury, and Carmelo E. Majorana. The control of upgrade activities for long tunnels by an intelligent system. In B.H.V. Topping, editor, Proceedings Of The Seventh International Conference On The Application Of Artificial Intelligence To Civil And Structural Engineering. Civil-Comp Press, 2003. [2] P. Katranuschkov, A. Gehre. An ontology framework to access IFC model data. ITcon Vol. 8 (2003), Special Issue eWork and eBusiness , pg. 413-437.

2

Abstract In this paper we provide the description of a formal ontology for the description of long tunnels. This object is used in the upgrade procedure of long tunnels on the web.

Keywords: Intelligent System, Formal Ontology, Structural Engineering, Long Tunnels, Upgrade Procedures.

1

Introduction

In this paper we shall illustrates one of the core parts of the ontological layer of an application, which is under development, whose purpose is to support upgrading of long tunnels. This application will be an ontology-driven web decision support system. The reasons for supporting this process in such a semi-automatic way are many: • The aspects that have to be considered when applying an upgrade procedure are so many that in practice it is impossible to manage them without an automated support; • The number of involved experts is high, and they usually do not speak the same technical language, and often do not speak the same natural language as well, therefore we need a support for a a unified model of knowledge, that will necessarily be a web-based one, due to the geographic spread of experts; • It is not possible to share data of different experts without a direct support of unified processing service. Many of the processing utilities presently employed or under development have proprietary input and output data formats, forcing 1

therefore the experts to transfer data to each other either through complex semiautomated translation processes or by means of human understanding, often supported by printed data sheets. In this paper we illustrate the conceptual organisation of data describing tunnels. The aspects included in this description are fundamentally five: the functional and prescriptive requirements of tunnels, the tunnel safety features, the tunnel structure and geometry, the data description of Mass-Heat Flow analysis, the notion of documents to be deployed in the IntelliTun application. For all the above mentioned aspects, the ontological perspective upon tunnels we have undertaken, is essentially based on the needs of the application, and in the sense which is proposed in the current literature about formal ontology is a task-oriented ontology. This justifies some technicalities of the structure which are not needed in a more general and less application-oriented specification.

2

Related Work

It is quite well established in recent investigation on Information Systems, that formal ontologies are a crucial problem to deal with, and in fact they received a lot of attention in several different communities, such as knowledge management, knowledge engineering, natural language processing, intelligent information integration, and so on [11]. Ontologies have been developed in Artificial Intelligence to facilitate knowledge sharing and reuse. The viewpoint we adopt here is taken from the general considerations on the use of philosophical issues in Artificial Intelligence: “the systematic, formal, axiomatic development of the logic of all forms and modes of being” [12]. Another commonly accepted definition is that an ontology is an explicit specification of a shared conceptualization that holds in a particular context. The actual topic of ontology is one of those themes that epistemology (theories on knowledge) dealt with in philosophical studies of Parmenides, Heraclitus, Plato, Aristotle, Kant, Leibnitz, Wittgenstein, and others. Ontologies define the kind of things that exist in the world and, possibly, in an application domain. In other words, an ontology provides an explicit conceptualization which describes semantics of data, providing a shared and common understanding of a domain. From an AI perspective we can say that: ”. . . ontology is a formal explicit specification of a shared conceptualization. Conceptualization refers to an abstract model of phenomena in the world by having identified the relevant concepts of those phenomena. Explicit means that the type of concepts used, and the constraints on their use are explicitly defined. Formal refers to the fact that the ontology should be machine readable. Shared reflect that ontology should capture consensual knowledge accepted by the communities[13].” 2

And moreover: ”An ontology may take a variety of forms, but necessarily it will include a vocabulary of terms, and some specification of their meaning. This includes definition and an indication of how concepts are inter-related which collectively impose a structure on the domain and constrain the possible interpretation of terms” [14] Nowadays, ontologies • are used to allow communication among people and heterogeneous and widely spread application systems; • are implied in projects as a conceptual model, to enable a content-base access on corporate knowledge memories, knowledge bases, archives; • allow agents to understand each other when they need to interact, communicate and negotiate meanings; • refer to a common piece of information and share common understanding of the information structure. The Description of Tunnel Safety Features is obtained as a synthesis of some basic documents, in particular: • Prescriptive or recommending documents: – The Official Recommendation of the Special Commission of the United Nations devoted to safety in Tunnels, whose chairman is Michel Egger and vice-chairman is Didier Lacroix, whose objective is to provide general recommendation to the governments of member states in order to establish laws and regulations that are uniform and meaningful in engineering terms [3]; – The official Manual of Road Tunnels Traffic and Fire Safety of the Norwegian Traffic Society [10]; – The recommendation of the TNO Consortium about safety features for the Netherlands [9]; • Technical document: – A paper about the Norwegian Road Tunnels [1]; – Some papers about fire safety in road tunnels related to the UPTUN project [6, 5]. 3

More generally speaking we have employed general notions of Formal Ontology as can be found in [7] and [4]. Moreover we shall employ the foundational methodology proposed by Uschold [8]. It is quite well established in recent investigation on Information Systems, that formal ontologies are a crucial problem to deal with, and in fact they received a lot of attention in several different communities, such as knowledge management, knowledge engineering, natural language processing, intelligent information integration, and so on [11]. Ontologies have been developed in Artificial Intelligence to facilitate knowledge sharing and reuse. The viewpoint we adopt here is taken from the general considerations on the use of philosophical issues in Artificial Intelligence: “the systematic, formal, axiomatic development of the logic of all forms and modes of being” [12]. Another commonly accepted definition is that an ontology is an explicit specification of a shared conceptualization that holds in a particular context. The actual topic of ontology is one of those themes that epistemology (theories on knowledge) dealt with in philosophical studies of Parmenides, Heraclitus, Plato, Aristotle, Kant, Leibnitz, Wittgenstein, and others. Ontologies define the kind of things that exist in the world and, possibly, in an application domain. In other words, an ontology provides an explicit conceptualization which describes semantics of data, providing a shared and common understanding of a domain. From an AI perspective we can say that: ”. . . ontology is a formal explicit specification of a shared conceptualization. Conceptualization refers to an abstract model of phenomena in the world by having identified the relevant concepts of those phenomena. Explicit means that the type of concepts used, and the constraints on their use are explicitly defined. Formal refers to the fact that the ontology should be machine readable. Shared reflect that ontology should capture consensual knowledge accepted by the communities[13].” And moreover: ”An ontology may take a variety of forms, but necessarily it will include a vocabulary of terms, and some specification of their meaning. This includes definition and an indication of how concepts are inter-related which collectively impose a structure on the domain and constrain the possible interpretation of terms” [14] Nowadays, ontologies • are used to allow communication among people and heterogeneous and widely spread application systems; • are implied in projects as a conceptual model, to enable a content-base access on corporate knowledge memories, knowledge bases, archives; 4

• allow agents to understand each other when they need to interact, communicate and negotiate meanings; • refer to a common piece of information and share common understanding of the information structure.

3

The structure of OntoTunnel

In this section we illustrate, majorly by describing a sequence of screenshot of the ontological layer of IntelliTun the approach we adopted in the production of the reference ontology OntoTunnel. In Figure 1 we provide the taxonomic organization of the safety features of tunnels. In particular, these features have been inventoried in a reference document of the UPTUN Project, named Inventory of Tunnel Safety Features, which is a technical report of the EU and is briefly illustrated [2]. The current reference individuates the following major classes: • Prevention methods • Detection/Monitoring systems • Suppression systems • Ventilation systems • Illumination systems • Communication systems • Evacuation systems and procedures • Structures • Emergency Infrastructures More specifically, for instance, prevention methods are: • Reduced speed limits • Speed controls • Increased distance between vehicles • Prohibited lane change and overtaking • Straight or gently curved tunnels 5

Figure 1: Tunnel safety features as implemented in Ontotunnel. • Gently sloped roadways • Marking with acoustic effect • Road surface (friction) • Periodic inspections of structures • Periodic inspections of technical equipment • Maintenance • Limits for vehicles dimensions • Hazardous materials and liquids prohibited • Habilitation for transport of hazardous materials • Habilitation for transport of high calorific power materials • Escorts for hazardous materials • Escorts for oversized vehicles • Trucks restricted to one lane • Portal inspections • Overweight indicators In Figure 2 two further concepts of tunnel ontology are specified: location, and the classification of tunnels. A location is a geographic/geometric position either in a GPS coordinates or as a numeric value measuring path length over a tunnel, described indeed as an open polygonal. Tunnels are classified in five ways: 6

Figure 2: The top level concepts of Location and Tunnel, and their subclasses as implemented in Ontotunnel. • Based upon their function (road, rail, metro tunnels); • Based upon their geometric characteristics with respect to the exterior (open, closed); • Based upon the relation between their usage and their shape (singular mileage, double mileage, separated); • Based upon their shape (rectangular, bell-shaped); • Based on their geological location (immersed, cut-and-cover). The upon mentioned ways do not combine to each other freely, for instance metro tunnel are necessarily separated). The ontology takes into account these things by minimising the admissible combination, committing to the ALARA approach (As Low As Reasonably Achieveable). Figure 3 pictorially illustrates the material and mass-heat flow concepts. MassHeat flow is the conceptual category under which Fire Engineers classify the behavior of structures in presence of accidents. The description of these events is included in the ontological layer, because the analysis of Mass-Heat Flow behaviour for fire scenarios is a discriminating analysis in decision-making activities of tunnel upgrade, and therefore needed in this application-driven ontological layer. Related to this aspect of accident analysis, we include the concept describing traffic aspects (in Figure 4), as well as the classification of those hazardous goods that can be transported through a tunnel (in Figure 5) and the classification of emergencies (in Figure 6). 7

Figure 3: Materials and Mass-Heat flow concepts.

Figure 4: Traffic related concepts.

8

Figure 5: Hazardous goods that can be transported through tunnels.

Figure 6: Emergencies in tunnels.

9

Acknowledgments We gratefully thank the European Community, under Grant GRD1-2001-40739 (the UPTUN Project), for supporting the investigation which we have documented in this paper. The UPTUN-project Cost-effective, sustainable and innovative upgrading methods for fire safety in existing tunnels is being carried out in the framework of the Competitive and Sustainable Growth Programme (project GRD1-2001-40739, Contract G1RD-CT-2002-0766), with a financial contribution of the European Community. We would also like to thank Michele Albrigo for his very useful help in both the development of the prototypes and their description.

References [1] Finn H. Amundsen, Guro Ranes, and P˚al Melvær. Studies on norwegian road tunnels. Technical report, Division of Transport Analysis, Norwegian Public Roads Administration, Oslo, March 1997. [2] Matteo Cristani. An ontology of tunnel safety features. Research Report RR 10/2003, Dipartimento di Informatica, Universit´a di Verona, 2003. [3] Jos´e Capel Ferrer, Michel Egger, and Didier Lacroix. Recommendations of the group of experts on safety in road tunnels: Final report. Technical report, United Nations - Economic and Social Council - Economic Commision for Europe Inland Transport Committee - Ad hoc Multidisciplinary Group of Experts on Safety in Tunnels, 2001. [4] N. Guarino. Formal Ontology, Conceptual Analysis and Knowledge Representation. International Journal of Human and Computer Studies, 43(5/6):625–640, 1995. [5] G.A. Khoury. Active and passive fire safety in tunnels. Industrial Fire Journal, March 2003. [6] G.A. Khoury. Eu tunnel fire safety action. Tunnels and Tunnelling International, April 2003. [7] John F. Sowa. Knowledge Representation:Logical, Philosophical, and Computational Foundations. Brooks Cole Publishing Co., Pacific Grove, CA, 2000. [8] M. Uschold and M. Gr¨uninger. Ontologies: Principles, methods and applications. Knowledge Engineering Review, 11(2):93–137, 1996. [9] Various. Proposal for safety criteria in road tunnels in the netherlands. TNOMEP R 2002/681, TNO, Amsterdam, 2002. 10

[10] Various. HB 021 Road Tunnels Traffic and Fire Safety. Norwegian Road Traffic Society, 2003. [11] D. Fensel, Ontologies: A silver bullet for knowledge management and electronic commerce, Springer, 2000. [12] Cocchiarella, N.B. Formal Ontology. In H. Burkhardt and B. Smith (eds.), Handbook of Metaphysics and Ontology. Philosophia Verlag, Munich, 1991. [13] T. R. Gruber, A translation approach to portable ontology specifications, Knowledge Acquisition 5 (1998), 199–220. [14] R. Jasper and M. Ushold, A Framework for Understanding and Classifying Ontology Applications.

11