Application of Ontology in Emergency Plan Management of Metro Operation Procedia Engineering 164 ( 2016 ) 158 – 165 1877 7058 © 2016 The Authors Published by Elsevier Ltd This is an open access articl[.]
Available online at www.sciencedirect.com ScienceDirect Procedia Engineering 164 (2016) 158 – 165 Creative Construction Conference 2016, CCC 2016, 25-28 June 2016 Application of ontology in emergency plan management of metro operation Hanbin Luo , Xiaofan Peng*, Botao Zhong Huazhong University of Science and Technology, Wuhan 430074, P.R China Abstract Emergency plans play a key role in the emergency management of metro operation Well-prepared emergency response supported by plans can greatly mitigate the significant impact of metro incidents However, most emergency plans remain as plain-text documents, which make it difficult to conduct efficient administrative work such as plan generation, documentation and maintenance Operational use of plans such as rapid knowledge retrieval and acquisition cannot be performed as well, affecting emergency training during the preparation process as well as plan review at the scene of the accident Additionally, the knowledge codified in emergency plans is mostly depicted by texts, which are not vivid and intuitive enough to clearly convey the instructions of response procedures and relative information In this paper, an ontology-based knowledge management method is proposed, and a unified and formalized plan repository is built to facilitate the efficient administrative and operational use of emergency plans BIM technology is applied to provide realistic visualization of the plan knowledge for better understanding A prototype of emergency plan training system, which integrates BIM and the ontology-based plan repository, has been developed to demonstrate the feasibility and effectiveness of the method A case study illustrates the knowledge management process and shows how staff training can benefit from the system 2016The TheAuthors Authors Published by Elsevier Ltd is an open access article under the CC BY-NC-ND license © 2016 © Published by Elsevier Ltd This Peer-review under responsibility of the organizing committee of the Creative Construction Conference 2016 (http://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review under responsibility of the organizing committee of the Creative Construction Conference 2016 Keywords: Emergency plan; Metro operation; Ontology; Training system Introduction With the characteristic of high speed, punctuality and environmental protection, metro has gained its popularity in China recently However, huge casualties or economic loss can be caused when metro incidents happen during * Corresponding author Tel.: +86 13986079085; fax: +0-000-000-0000 E-mail address: pxf_angeling@163.com 1877-7058 © 2016 The Authors Published by Elsevier Ltd This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/) Peer-review under responsibility of the organizing committee of the Creative Construction Conference 2016 doi:10.1016/j.proeng.2016.11.605 Hanbin Luo et al / Procedia Engineering 164 (2016) 158 – 165 operation, such as station fire, train breakdown and terrorist attack The increasing scales of metro construction as well as the booming passenger flow have put forward higher demand for the emergency management of metro incidents To better deal with these emergencies, Chinese government has issued national guidance to support the development of emergency plan systems in different cities Hence, a large amount of emergency plans covering all possible incidents and all operation areas have been developed by local metro operation companies The plans serve as a manual that mainly describes procedures for dealing with all kinds of emergencies [1] These plans will be learnt during staffs’ professional training to improve their proficiency of the knowledge so that they can act quickly at the scene When metro incidents happen, the corresponding emergency plan will be initiated and support the onsite response work The plans can also be improved and updated based on the implementation effect However, more than 80% emergency plans of metro operation in China are plain-text documents or rigid electronic files without semantic meaning, which makes it difficult to conduct efficient administrative work such as plan generation, documentation and maintenance It also causes inconvenience in the use of plans Since large amounts of knowledge fragments are scattered among various files without organization, rapid knowledge retrieval and acquisition can’t be easily achieved, considerably affecting training effect for staffs as well as on-site knowledge support for decision makers Thus, it is essential to organize and represent the plan knowledge in a structured and coherent manner to promote the efficiency both in administrative and operational use of emergency plans Meanwhile, the knowledge codified in emergency plans are mostly depicted by texts, and generalized descriptions of response procedures can’t effectively convey the information about how to respond in a specific situation, which leads to the lack of pertinence and effectiveness of emergency plans in real situations Therefore, combining response procedures with its associated environmental information is important to enhance the applicability of plans, and a more visualized method of knowledge display is needed to extend the capacity of metro staffs to better understand and memorize plans In this work, ontology has been introduced for knowledge management support It can provide coherent and accurate conceptualization of domain knowledge [2] to integrate the relevant information in a common manner Also, the knowledge can be presented in a standard and semantic way to avoid the ambiguity of natural language and enhance computer-aided information processing Hence, to overcome the problems of traditional file management of emergency plans mentioned above, an ontology-based knowledge management has been proposed to improve the efficiency both in administrative work and operational use of plans A prototype of emergency plan training system for metro staffs, which combines the ontology-based plan repository and BIM, has been developed to support efficient emergency plan training process for metro staffs A case study illustrates the knowledge management process and shows how staff training can benefit from the system The remainder of this paper is organized as follows: In Section 2, we briefly discuss the related work in information management of emergency plans and BIM application in emergency simulation Section presents the design and construction of the ontology of emergency plans for metro operation Section discusses the establishment of the prototype system for metro staff training Section illustrates an example of applications Finally in Section we conclude the work, discuss the prospect of the ontology-based training system for emergency responders as well as future study directions Research review To improve the management efficiency of emergency plans, various and individual information management systems have been developed for different research aims Zhang, Li and Wang [3] have formed a digital plan database, based on which an emergency decision support platform was developed to support procedure retrieval Dong, Li and Xu [4] have developed an information management system to enhance the automatic management of emergency plans such as adding, deleting, modifying and inquiry of plans Canós et.al [5] introduced SAGA, a framework designed to provide support to the full lifecycle of emergency plan management and use, especially the administrative aspects Ontology is also introduced to solve the semantic problem Mejri [6] introduced ontology for the analysis of information function during the entire workflow of emergency management Yu and Wang [7] also constructed an emergency ontology to reorganize the knowledge pieces from emergency plans for decision-making support Wang, Yang and Dong [8] have built an emergency plan system ontology to promote the communication 159 160 Hanbin Luo et al / Procedia Engineering 164 (2016) 158 – 165 and sharing between different plan systems Most researches study it from the administrative perspective, and ontology construction for operational use of emergency plans is seldom considered To enhance the applicability of plan knowledge by integrating contextual information, Canós, Alonso and Jaen [1] have proposed the concept of hypermedia plan which integrated the plan information with a 3D model of the station on one screen Díez et.al [9] presents a computer-based collaborative environment to share team members’ view to support team communication during the post-hoc review of emergency plan The booming of BIM technology in recent years has brought information visualization to a higher level Chen and Chu [10] have developed an approach to retrieve the building geometry from the BIM model for graph construction to support route planning The similar method was used by Li et.al [11] to support building fire emergency response operation Furthermore, virtual reality and computer gaming technology have been combined with BIM by Ruppel and Schatz [12] to provide realistic visualization of emergency scenarios for fire fighters to interact with it However, BIM-based knowledge display for emergency plan is seldom studied Hence, this research has built a standard and semantic ontology-based knowledge database of emergency plans to improve the efficiency in its administrative work and operational use It serves as the foundation of intelligent systems and applications Also, BIM is applied to provide specific and intuitive knowledge display of emergency plans The ontology-based knowledge management of emergency plans will be firstly presented in the next section Ontology-based knowledge management of emergency plans The most widely recognized definition of ontology is given by Gruber [13], “Ontology is an explicit specification of a conceptualization.” It provides a way to move from a document-oriented view of knowledge management to a content-oriented view [14], and to build a common understanding of domain knowledge by identifying glossary and relations between them Another characteristic of ontology is formalization, which was mentioned by Studer et.al [15] to highlight its machine-readable feature The ontology-based knowledge management covers four steps, as shown below, and detailed explanation followed x x x x Knowledge Identification Knowledge Extraction Knowledge Modeling Knowledge Representation Firstly, sources such as national guidance and emergency plans in several cities will be consulted to determine the knowledge domain, and to identify the necessary information that should be included in the plan, such as response organizations, procedures and rescue resources Secondly, a common language and principles shall be developed to standardize the information Unified concepts and relations will be extracted to conceptualize the identified information of emergency plans Then, an ontology-based knowledge model can be build based on these concepts and relations The model also serves as a shared template to organize the vast knowledge fragments among different plans and to build a coherent knowledge repository of plans Finally, a representation schema are needed to represent the knowledge in a formalized and explicit way A common ontology language with machine-readable feature will be used to achieve further computer-aided information processing After that, a structured, standardized and formalized knowledge repository of emergency plans for metro operation can be constructed The administrative work can be easily achieved such as generating a new plan with reusable and shared template, making fraction updating without going through the whole paper and performing exact search for plan reviewing by different actors As for its operational use, the ontology-based knowledge representation can promote computer-aided emergency management work such as high-quality staff training or timely knowledge support for decision making at the scene 161 Hanbin Luo et al / Procedia Engineering 164 (2016) 158 – 165 (PHUJHQF\ 3ODQ (PHUJHQF\3ODQ ,QLWLDWLRQ 6WDWLRQ IHDWXUH 5RDG (PHUJHQF\ ,QGH[ 1HDUE\ %XLOGLQJ /LQH 6WDWLRQ +DOO ,QFLGHQW &KDUDFWHULVWLF 3ODWIRUP 6WDWLRQ 5HVSRQVH 3URFHGXUH 0DQDJHPHQW 5RRP 5HVRXUFH $PRXQW 2UJDQL]DWLRQ 3URFHGXUH 6WDWLRQ 6WDII ,QFLGHQW 7\SH ([LW 5HVFXH 5HVRXUFH 'HSDUWPHQW 0DQDJHU 3DVVHQJHU $WWHQGDQW *HQHUDO 0DQDJHU 5HVRXUFH 7\SH 5HVRXUFH /RFDWLRQ 6WDWLRQ PDQDJHU 6WDWLRQ $WWHQGDQW 6WDWLRQ 2SHUDWRU Fig Knowledge model of emergency plans for metro operation 3.1 Knowledge modeling of emergency plan After a study of national guidance and emergency plans of several cities, key concepts and relations are captured and extracted from these scattered sources for knowledge modeling, and a common knowledge model of emergency plan has been built, as shown in Fig.1 Four core parts of a complete emergency plan have been identified as below x x x x Station Feature Incident Characteristic Response Procedure Rescue Resource Station feature includes basic information about physical structures, spatial arrangement and surrounding environment of metro stations Incident characteristic is to describe the feature of one kind of emergency such as incident type and emergency index The index is used to determine whether the emergency will happen or not and measure the severity degree of the situation As the most import part of plans, response procedure provides detailed action guidance during an emergency, and organization presents the party who will perform the actions In this research, station staffs are the major concerned objects Rescue resources such as emergency lamps and gas masks are needed to help with the action performance Metro staffs at the station should be familiar with their resource type, amount, location and availability so that they can find the needed one timely The most essential relations between these concepts are constructed based on the initiation mechanism of an emergency plan, as shown in the dotted rectangle According to the station feature and incident characteristic of a real emergency, relative response procedures can be determined 3.2 Knowledge representation Based on the proposed knowledge model, a prototype ontology of emergency plans for metro operation (OntoEP4MO) can be built The ontology has been implemented by using protégé, which is a software providing an easy and flexible interface to create ontology The basic elements of the ontology can be constructed according to the concepts and relations proposed above Concepts become classes, while sub-concepts become subclasses in the form of a superclass-subclass hierarchy The simple relations between each concept change into properties The real-case information can be stored as instances of the corresponding classes Moreover, the complex relations that can’t be described clearly by properties will be established into rules, e.g the mechanism of plan initiation The construction of rules and instances will be illustrated in details in Section based on a case study Fig.2 (a) shows a part of the protégé interface of the prototype ontology 162 Hanbin Luo et al / Procedia Engineering 164 (2016) 158 – 165 Fig (a) Protégé Interface of Onto-EP4MO; (b) Partial OWL File To facilitate computer-aided information processing, the knowledge needs to be presented in a formalized and explicit way with machine-readable language OWL, as a common ontology language, has been used to define and describe ontology elements By codifying collected knowledge through a formal semantics [16], it can facilitate the interoperability with other information management platforms Moreover, an OWL DL reason tool called Pellet is used as well to verify the consistency of ontology classes [17], and a part of OWL file is shown in Fig.2.(b) For now, a structured and formalized knowledge repository of emergency plans for metro operation is established, based on which intelligent systems and applications can be developed to improve the efficiency during emergency management process Emergency plan training system The ontology-based plan repository can be applied to prepare metro staffs in their professional training, emergency drills or provide knowledge support for on-site command and relief work In this research, a prototype of emergency training system for metro staffs has been developed Considering the poor effect of text-based learning method, BIM is used to provide specific and realistic visualization of plan knowledge 4.1 System architecture The architecture of the system is shown in Fig.3 In the data layer, the ontology-based plan repository is to provide the plan knowledge, while BIM model is to provide the geometric and functional information of the station in a more accurate and virtual way The BIM-based simulations of response procedures can be created based on Unity3D and connected to the relative plan knowledge In the function layer, four function modules of the system relative to four core parts of an emergency plan have been built to perform the plan knowledge learning For example, in the function module of “Station Feature”, metro staffs can have a good understanding of their working station by operating on the BIM model Also, they can have access to the system by using laptops, pads and smart phones, as shown in the operation layer Fig System architecture of emergency plan training system Hanbin Luo et al / Procedia Engineering 164 (2016) 158 – 165 4.2 BIM-based training environment As one of the most promising development in AEC industry, BIM can provide accurate virtual models in a digital way to support the management activities through the whole lifecycle of the building [18] Semantically rich information including all geometric and functional properties related to the facility can be integrated into the BIM model [19], improving visualization as well as the understanding of the relevant management work Instead of complicated 2D drawings, the BIM model of the station can be provided for staffs to get a quick knowledge about the station feature Detailed response procedures including walking routes and operation instructions can be shown accurately by BIM-based simulations to improve staffs’ cognition Therefore, by using BIM, the knowledge of emergency plans can be displayed in a more specific and visualized way to improve the training effect To build a BIM database for metro stations, The Autodesk Revit, as a design and documentation platform that supports design, drawings and schedules of the building projects [20], has been used for modeling work Case study To demonstrate the effectiveness of the ontology-based emergency plan repository and the feasibility of the training system, one kind of emergency plan prepared for the emergency incident named “Passenger Flow Surge” is chosen as examples The incident describes a situation where the number of passengers who stay on the station hall or the platform exceeds the capacity of the station, and may cause overcrowding and trample The plan is to maintain order and prevent jams by easing passenger traffic By using ontology, the plan knowledge written in files are stored in the repository and processed efficiently by computers The BIM model of the metro station named “ZhongNan Road Station” is built to provide a more accurate and vivid training environment Then, plan training for metro staffs can be achieved based on the system 5.1 Knowledge management of emergency plan On the basis of Ont-EP4MO, the plan knowledge prepared for Passenger Flow Surge are stored in the plan repository An exemplified illustration of how to create ontology instances according to specific plan knowledge is shown in Fig.4 In this case, the severity of the emergency is classified into three levels, resulting in three kinds of incident types The passenger number on the station hall and the platform are the emergency indexes to evaluate the severity degree and determine the incident type The thresholds of the passenger number for each level have been set up Aiming at different incident types, targeted response procedures conducted by station staffs are determined Here, only the response procedures for the incident type of “Passenger flow Surge Level 3” are shown due to limited space Such complex knowledge content like rules for level evaluation and procedure determination are core parts in the initiation mechanism of an emergency plan, as mentioned before They can’t be represented by only creating ontology properties Hence, SWRL (Semantic Web Rule Language) [21] is introduced here to establish complicated rules based on ontology elements, as shown in Fig.5.(a) For example, “EmergencyCase(?) ġ hasStationName(?x, ZhongNan_Road_Station) ġ hasPlatformPassengerNumber(?x, ?y) ġ swrlb:greaterThan(?y, 6700) ė hasIncidentType(?x, Passenger_Flow_Surge_Level3)” is the rule of level evaluation, which means if the passenger number on the platform of ZhongNan Road station exceeds 6700, the incident type named “Passenger_Flow_Surge_Level3” shall be reasoned The rule of “EmergencyCase(?x) ġ hasIncidentType(?x, Passenger_Flow_Surge_Level3) ġ hasPosition(?x, Station_Attendant_B) ė hasProcedure(?x, direct_the_crowd_at_the_escalator)” means when the incident type is Passenger Flow Surge Level 3, the response procedure of station attendant B should be directing the crowd at the escalator 163 164 Hanbin Luo et al / Procedia Engineering 164 (2016) 158 – 165 Fig Knowledge modeling process of emergency plan for passenger flow surge Fig (a) SWRL rules construction and knowledge reasoning; (b) BIM-based simulation of response procedures 5.2 Emergency plan training In this case, the number of passengers stay on the platform was set as 6750, exceeding the threshold of level 3, so the incident type will be reasoned as “Passenger_Flow_Surge_Level3” The response procedure “direct_the_crowd_at_the_escalator” will be acquired for the staff named “Station_Attendant_B” In the function module of “Response Procedures” of the system, the related BIM-based simulations showing how to direct the crowd at the escalator in ZhongNan Road station can be displayed for staff learning, as shown in Fig.5.(b) Since the knowledge has been stored and managed in a structured and semantic manner, such knowledge processing can be efficiently achieved by computers Compared with looking for instructions in numerous files, it saves a lot time and efforts, greatly improving the training equality Conclusion and future work Efficient emergency response is crucial and lies at the heart of the emergency management of metro incidents Considering the inefficient management of paper-based emergency plans and the follow-up problems in its operational use, an ontology-based knowledge management of emergency plans for metro operation has been proposed to organize and represent the plan k in a standard and semantic way, enhancing the efficiency in both administrative and operational use of emergency plans With this approach, the automation degree of emergency Hanbin Luo et al / Procedia Engineering 164 (2016) 158 – 165 plan management will be advanced, and a collaborative environment for different parties involved in the whole lifecycle management of plans can be provided for information sharing and reuse The prototype of emergency plan training system has guaranteed a promising pathway to make the emergency management of metro operation more efficiently A wide generalization of more intelligent systems or applications can be foreseen However, two limitations are identified below: since the ontology was constructed only based on existed emergency plans, other important knowledge that are not covered in the plan may be missed Interviews of professionals in the field of emergency management should be conducted for continuous refinement and evaluation of the ontology The proposed training system should be testified by more station staffs to ensure its operability in practical applications Further development of the emergency plan training system is the research direction Acknowledgements This research was supported in part by a major project of The National Social Science Key Fund of China (Grant No.13&ZD175), the National Natural Science Foundation of China (Grant No 71301059) and the Fundamental Research Funds for the Central Universities (Grant No 0118240099) The authors also would like to thank anonymous reviewers who gave valuable suggestion that has helped to improve the quality of the manuscript References [1] Canós, J.H., G Alonso and J Jaén, A multimedia approach to the efficient implementation and use of emergency plans MultiMedia, IEEE, 2004 11(3): p 106-110 [2] Galton, A and M Worboys An ontology of information for emergency management In 8th International Conference on Information Systems for Crisis Response and Management (ISCRAM 2011), Lisbon 2011 [3] Zhang, M., P Li and F Wang The emergency decision support platform of urban rail transit based on information sharing and digitalized plan In Proceedings of the Tenth International Conference of Chinese Transportation Professionals 2010 [4] Dong, H., F Li and R Xu Design of Emergency Preplan Exercise Management Information System for URT in CICTP 2012@ sMultimodal Transportation SystemsಧConvenient, Safe, Cost-Effective, Efficient 2012: ASCE [5] Canós, J.H., et al., Improving emergency plans management with SAGA Technological Forecasting and Social Change, 2013 80(9): p 1868-1876 [6] Mejri, O Information conceptualization for emergency management In Towards a Service-Based Internet ServiceWave 2010 Workshops 2011: Springer [7] Yu, K., Q Wang and L Rong Emergency Ontology construction in emergency decision support system In Service Operations and Logistics, and Informatics, 2008 IEEE/SOLI 2008 IEEE International Conference on 2008: IEEE [8] WenJun, W., D CunXiang and Y Peng Ontology modeling of emergency plan systems In Fuzzy Systems and Knowledge Discovery, 2009 FSKD'09 Sixth International Conference on 2009: IEEE [9] Díez, D., et al., Sharing your view: A distributed user interface approach for reviewing emergency plans International Journal of HumanComputer Studies, 2014 72(1): p 126-139 [10] Chen, A.Y and J.C Chu, TDVRP and BIM Integrated Approach for In-Building Emergency Rescue Routing Journal of Computing in Civil Engineering, 2015: p C4015003 [11] Li, N., et al., A BIM centered indoor localization algorithm to support building fire emergency response operations Automation in Construction, 2014 42: p 78-89 [12] Wang, B., et al., BIM based virtual environment for fire emergency evacuation The Scientific World Journal, 2014 2014 [13] Gruber, T.R., A translation approach to portable ontology specifications Knowledge acquisition, 1993 5(2): p 199-220 [14] Staab, S., et al., Knowledge processes and ontologies IEEE Intelligent systems, 2001(1): p 26-34 [15] Studer, R., V.R Benjamins and D Fensel, Knowledge engineering: principles and methods Data & knowledge engineering, 1998 25(1): p 161-197 [16] Onorati, T., et al., Modeling an ontology on accessible evacuation routes for emergencies Expert Systems with Applications, 2014 41(16): p 7124-7134 [17] Malizia, A., et al., CAP-ONES: an emergency notification system for all International Journal of Emergency Management, 2009 6(3-4): p 302-316 [18] Eastman, C., et al., BIM handbook: A guide to building information modeling for owners, managers, designers, engineers and contractors 2011: John Wiley & Sons [19] Ding, L., Y Zhou and B Akinci, Building Information Modeling (BIM) application framework: The process of expanding from 3D to computable nD Automation in Construction, 2014 46: p 82-93 [20] Krygiel, E., P Read and J Vandezande, Mastering Autodesk Revit Architecture 2011 2010: John Wiley & Sons [21] Consortium, W.W.W., SWRL: a semantic web rule language combining OWL and RuleML W3C member submission, 2004 165 ... the heart of the emergency management of metro incidents Considering the inefficient management of paper-based emergency plans and the follow-up problems in its operational use, an ontology- based... the ontology of emergency plans for metro operation Section discusses the establishment of the prototype system for metro staff training Section illustrates an example of applications Finally in. .. of plans A prototype of emergency plan training system for metro staffs, which combines the ontology- based plan repository and BIM, has been developed to support efficient emergency plan training